Roles of Platelet-Derived Growth Factor-C and Early Growth Response-1 in Vascular Smooth Muscle Cells

University of New South Wales Faculty of Medicine Centre for Vascular Research

Roles of Platelet-derived Growth Factor-C and Early Growth Response-1 in Vascular Smooth Muscle Cells

A thesis presented for the degree of

Doctor of Philosophy

by YANEDTH ESTELLA SANCHEZ-GUERRERO

2010 Originality Statement

‘I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project's design and conception or in style, presentation and linguistic expression is acknowledged.’

Signed ………………………………………… Yanedth Estella Sanchez-Guerrero

Date …………………………………………

‘I hereby grant the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all proprietary rights, such as patent rights. I also retain the right the use in future works (such as articles or books) all or part of thesis or dissertation.

I authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstracts International.

I have either used no substantial portions of copyright material in my thesis or I have obtained permission to use copyright material; where permission has not been granted I have applied/will apply for a partial restriction of the digital copy of my thesis or dissertation.’

Signed ………………………………………… Yanedth Estella Sanchez-Guerrero

Date …………………………………………

Acknowledgments

Levon, you are my role model. Thanks for giving me the opportunity to be part of your research group. As I have told you many times Levon: “it is my honour to be your student”.

Mary K and Fernando: you have been the best of the very best. Thanks for your patience and your willingness to answer my questions and helping me with this project.

Lucinda, Belinda and Margaret; I will never be able to pay you for your invaluable help. Thank you!

Lionel I thank you not just for your friendship but also for your encouragement throughout the years.

To everyone else in the laboratory and to my ex-colleagues: thanks for all your help.

Keith, you are such a wonderful person. Thanks for your continuous support. I also thank you for giving me a hand with the figures.

And the most important one: my gorgeous family, mum, dad and beautiful sisters, my “babies” and my brother in law: your love, encouragement, support, collaboration, have brought me to where I am now. Although not being together, you have had me on your thoughts.

Finally, GOD…..thanks for allowing me to accomplish what I always wanted: my PhD degree.

Publications

Sanchez-Guerrero E, Khachigian LM. Egr-1 induction by PDGF-BB involves

MMP/ADAM-dependent ERK1/2, PDGFRβ, and EGFR phosphorylation, and

PDGFRβ-EGFR heterodimerisation. Arteriosclerosis, Thrombosis, and Vascular

Biology. (Submitted).

Sanchez-Guerrero E, Midgely VC, Khachigian LM. (2008). Angiotensin II induction

of PDGF-C expression is mediated by AT1 receptor-dependent Egr-1 transactivation.

Nucleic Acids Res 36(6), 1941-51.

Prizes

Australian Vascular Biology Society Young Investigator/Travel Award.

16th International Vascular Biology Meeting, Los Angeles, 2010.

Dean’s list Award. School of Medical Sciences, Faculty of Medicine, UNSW. 2008

Conference Poster Presentation Abstracts

Sanchez-Guerrero E, Khachigian LM. Egr-1 induction by PDGF-BB involves

MMP/ADAM-dependent ERK1/2, PDGFRβ, and EGFR phosphorylation, and

PDGFRβ-EGFR heterodimerisation. The 18th Australian Vascular Biology Meeting,

Lorne, September 2010.

Sanchez-Guerrero E, Khachigian LM. Egr-1 induction by PDGF-BB involves

MMP/ADAM-dependent ERK1/2, PDGFRβ, and EGFR phosphorylation, and

PDGFRβ-EGFR heterodimerisation. The 16th International Vascular Biology Meeting,

Los Angeles, June 2010.

Sanchez-Guerrero E, Midgely VC, Khachigian LM. Angiotensin II induction of

PDGF-C expression is mediated by AT1 receptor-dependent Egr-1 transactivation.

The 15th International Vascular Biology Meeting, Sydney, June 2008.

Abstract

Smooth muscle cell (SMC) proliferation and migration are key pathophysiological processes regulating vascular remodelling and atherosclerosis. Members of the platelet- derived growth factor (PDGF) family have been identified as participants of such processes. Although the classical PDGF members, PDGF-A and PDGF-B, have been extensively investigated, the newer members, PDGF-C and PDGF-D, are just beginning to be explored. Although the role of PDGF-C in a number of pathophysiological conditions is being investigated, the molecular mechanisms behind its transcriptional activation are yet to be elucidated.

The first study in this thesis explored the transcriptional regulation of PDGF-C by

Angiotensin II (ATII) in SMCs. So far, no studies have linked ATII with PDGF-C expression. This study demonstrated for the first time that in neonatal SMCs ATII is capable of inducing PDGF-C via signalling through the ATII type 1 receptor (AT1R) and

Egr-1. We also identified a novel Egr-1 binding element in the PDGF-C promoter. Not only is PDGF-C up-regulation Egr-1 dependent but intriguingly is SMC subtype- dependent, underlying the heterogeneity of the SMC population.

The zinc-finger transcription factor Egr-1 serves as a key regulator of the inducible transcription of many genes implicated in the onset and progression of atherosclerosis.

However, understanding of the signalling mechanisms regulating its rapid induction after agonist exposure is poorly understood, beyond the involvement of the RAS/RAF/MEK/ERK or MKK/JNK pathways and serum-response element (SRE)- dependent transcription. In the second study, the mechanisms behind PDGF-BB induction of Egr-1 were examined. In neonatal WKY12-22 SMCs, it was discovered that

MMPs/ADAMs and EGFR play essential roles in signalling via PDGF-

BB/PDGFRβ/EGFR/ERK1/2/Egr-1. These findings also provide new mechanisms by which PDGFRβ can heterodimerise with EGFR, and emphasises the role of EGFR in

PDGFRβ-inducible downstream signaling.

Abbreviations

α alpha ADAM a disintegrin and a metalloproteinase ATII angiotensin II AT1R angiotensin II type 1 receptor AT2R angiotensin II type 2 receptor APS ammonium persulphate ASMC adult smooth muscle cells β beta bp base pairs cDNA complementary deoxyribonucleic acid ChIP chromatin immunoprecipitation cm2 square centimetres

CO2 carbon dioxide Cpm counts per minute dATP deoxyadenosine triphosphate dCTP deoxycytosine triphosphate dGTP deoxyguanine triphosphate dH2O deionised water DLR dual luciferase reporter DNA deoxyribonucleic acid dNTP deoxyribonucleoside triphosphate DTT dithiothreitol EC endothelial cells ECM extracellular matrix EDTA ethylenediaminetetraacetic acid EGFR epidermal growth factor receptor Egr-1 early growth response-1 ERK extracellular signal-regulated kinase EMSA eletromobility shift assay EtBr ethidium bromide FCS foetal calf serum FGF-2 fibroblast growth factor GPCR G-protein-coupled receptors

H2O2 hydrogen peroxide HDAC histone deacetylases h hour IL-1β interleukin1-β IB immunoblotting IP immunoprecipitation κ kappa kDa kilo Daltons JNK c-jun N-terminal kinase L litre M molar mM millimolar MAPK mitogen-activated protein kinases MEK mitogen-activated kinase/ERK kinase min minutes μg microgram μL microliter μM micromolar MMPs metalloproteinases mL milliliter mRNA messenger RNA PBS phosphate buffered saline PDGF platelet-derived growth factor PDGFRα platelet-derived growth factor receptor alpha PDGFRβ platelet-derived growth factor receptor beta PKC protein kinase C PMSF phenylmethylsulfanyl fluoride RNA ribonucleic acid ROS reactive oxygen species rpm revolutions per minute RPTK receptor protein tyrosine kinase RT room temperature RT-PCR reverse transcriptase polymerase chain reaction s seconds SDS sodium dodecyl sulphate Ser serine siRNA small interfering RNA SMCs smooth muscle cells Src sarcoma SH2 Src Homology 2-domain-containing Sp1 specificity protein-1 STAT signal transducers of transcription T75 75-cm2 flasks TGF-β transforming growth factor beta Thr threonine Tyr tyrosine TNF-α tumor necrosis factor alpha V volts VEGF vascular endothelial growth factor WKY wistar Kyoto WM waymouth’s Medium

Table of Contents

1 Introduction 1 1.1 Atherosclerosis 1 1.2 Atherogenesis 2 1.2.1 The artery wall 2 1.2.2 Initiation and progression of atherogenesis 3 1.2.3 Initiation and progression of atherogenesis: The hypothesis 5 1.2.3.1 Smooth muscle cells (SMCs) 7 1.2.3.2 Environmental cues 8 1.3 Platelet-derived growth factor (PDGF) family 9 1.3.1 Overview 9 1.3.2 PDGF family of growth factors 10 1.3.2.1 PDGF-A and PDGF-B: the classical PDGFs 14 1.3.2.1.1 PDGF-B 15 1.3.2.2 PDGF-C and PDGF-D: the new members 16 1.3.2.2.1 PDGF-C 17 1.3.2.2.1.1 PDGF-C and atherosclerosis 20 1.3.2.3 The PDGF receptors 21 1.3.2.4 Signalling via the PDGF receptors 25 1.3.2.4.1 Receptor kinase activity 25 1.3.2.4.2 Activation of signalling pathways 26 1.3.2.4.2.1 Mitogen-activated protein kinase (MAPK) cascade 29 1.4 Angiotensin II 32 1.4.1 ATII and CVD 35 1.5 Matrix metalloproteinases 38 1.5.1 Overview 38 1.5.2 Characteristics of the MMP family 38 1.5.3 Regulation of MMPs 43 1.5.3.1 Gene expression 43 1.5.3.2 Post translational modifications 44 1.5.3.3 Chromatin remodelling 44 1.5.3.4 Tissue inhibitors of MMPs 45 1.5.3.5 Pharmacological inhibitors of MMPs 47 1.5.4 ADAM family 50 1.5.4.1 ADAM-17 54 1.5.5 Epidermal growth factor receptor (EGFR) 56 1.5.5.1 EGFR ligands 58 1.5.5.1.1 EGFR ligands and atherosclerosis 59 1.5.5.2 EGFR activation by MMPs/ADAMs 62 1.5.5.3 G-protein coupled receptors (GPCR)-EGFR cross talk: triple membrane passing signaling 63 1.5.5.4 EGFR cross talk with PDGFR 68 1.6 Transcription factors 69 1.6.1 Early growth response factor (Egr-1) 69 1.6.2 Specificity protein (Sp1) 77 1.7 Aims and hypothesis of this thesis 80

2 Materials and Methods 82 2.1 Media, buffers and solutions 82 2.2 Cell culture 86 2.3 Gene expression analysis 87 2.3.1 Total RNA isolation 87 2.3.2 cDNA synthesis from total RNA 88 2.3.3 Semi-quantitative polymerase chain reaction (PCR) 89 2.3.4 Real-time PCR (qPCR) 90 2.4 Western blotting analysis 91 2.5 Statistical analysis 93 2.6 Densitometry 93 2.7 Other Methods 93

3 Angiotensin II-induction of Platelet-derived Growth Factor-C Expression is Mediated byAngiotensin II Receptor 1-dependent Egr-1transactivation 94 3.1 Introduction and Aim 94 3.2 Material and Methods 97 3.2.1 Chemicals 97 3.2.2 Cell culture 97 3.2.3 RNA preparation and reverse transcriptase reaction 97 3.2.4 Semi-quantitative and real-time PCR 98 3.2.5 Transient transfection 101 3.2.5.1 Plasmids 101 3.2.5.2 Transformation of competent cells 101 3.2.5.3 Maxi-prep: large scale DNA purification 102 3.2.5.4 Promoter-dependent expression experiments 103 3.2.5.5 Overexpression studies 105 3.2.5.6 DNAzyme transfection 106 3.2.6 Western blot analysis 106 3.2.7 Chromatin immunoprecipitation assay (ChIP) 108 3.2.8 Preparation of nuclear extracts 111 3.2.9 Electrophoretic mobility shift assay (EMSA) 111 3.2.10 Densitometry Analysis 112 3.2.11 Statistical Analysis 112 3.3 Results 113 3.3.1 ATII transiently induces PDGF-C mRNA expression in vascular SMC of neonatal origin 113 3.3.2 ATII stimulates PDGF-C promoter activity 115 3.3.3 Egr-1/Sp-1 binding site (-35/-1 bp) in the PDGF-C promoter is not required under ATII exposure 117 3.3.4 A novel Egr-1 binding site motif in the PDGF-C promoter (-543/-535) acts as a functional Egr-1-response element in WKY12-22 SMCs 121 3.3.5 Egr-1 and Sp1 interact and bind to the PDGF-C promoter 128 3.3.6 Egr-1 and Sp1 cooperatively transactivate the PDGF-C promoter 130 3.3.7 Egr1/Sp1 binding element (-543/-535) in the PDGF-C promoter is required for ATII-inducible PDGF-C transcriptional activation 132 3.3.8 Egr-1 DNAzyme blocks ATII-inducible PDGF-C expression 134

3.3.9 Egr-1 overexpression potentiates the activation of the PDGF-C promoter by ATII 137 3.3.10 ATII stimulates PDGF-C mRNA expression via Angiotensin II receptor 1 (AT1R) 139 3.3.11 Egr-1 induction of PDGF-C transcriptional activation is abrogated upon mutation of a putative Egr-1 motif (-543/-535) in the PDGF-C promoter 141 3.3.12 ATII differentially regulates PDGF-C in phenotypically distinct SMC subtypes 143 3.4 Discussion 145

4 Conclusion and Future directions: “Angiotensin II induction of Platelet-derived growth factor-C expression is mediated by Angiotensin II Receptor 1 dependent Egr-1 transactivation 150 4.1 Conclusions 150 4.1.1 ATII is confirmed to be an agonist of all four known PDGF ligand chains 150 4.1.2 ATII-induction of PDGF-C is controlled by Egr-1 and occurs through a novel Egr-1 binding site in the PDGF-C promoter requires cooperative interactions between Egr-1 and Sp-1 in neonatal WKY12-22 SMCs 151 4.1.3 ATII positively regulates PDGF-C expression in a SMC- subtype dependent manner 153 4.1.4 ATII positively regulates PDGF-C expression through AT1R 154

4.2 Future directions 157

4.2.1 Does ERK1/2 play a role in ATII-inducible PDGF-C expression? 157

4.2.2 Is Egr-1 and PDGF-C involved in ATII-induced migration of SMCs? 158 4.2.3 The role of Sp-1 on the PDGF-C promoter activation by ATII 158

4.2.4 Which other transcription factors mediate ATII-inducible PDGF-C expression? 160 4.2.5 Explore phenotypical differences between SMCs subtypes 160 4.2.6 Role of PDGF-C in atherogenesis 161

5 Egr-1 Induction by PDGF-BB Involves MMP/ADAM- dependent ERK1/2, PDGFRβ and EGFR Phosphorylation and PDGFRβ/EGFR Heterodimerisation 164 5.1 Introduction and Aim 164

5.2 Material and Methods 168

5.2.1 Chemicals 168

5.2.2 Cell culture 168

5.2.3 Total RNA preparation and cDNA synthesis 168 5.2.4 Real-time PCR 171 5.2.5 Co-immunoprecipitation 171 5.2.6 Western blot analysis 173 5.2.7 Small interfering RNA (siRNA) studies 175 5.2.8 Wound Scratch assay 175 5.2.9 Statistical and densitometry analyses 175 5.3 Results 176 5.3.1 PDGF-BB transiently induces Egr-1 expression in WKY12-22 SMCs and occurs via the ERK1/2 pathway 176 5.3.2 PDGF-BB induction of Egr-1 expression in WKY12-22 SMCs is MMP-dependent 179 5.3.3 PDGF-BB induction of Egr-1 expression in WKY12-22 SMCs may require other MMPs different from MMP-2, MMP-3 and MMP-9 189 5.3.4 Activation of MMPs with APMA, but not with Plasmin nor Urokinase-type plasminogen activator enhances PDGF-BB- induced Egr-1 expression in WKY12-22 SMCs 192 5.3.5 APMA rescues PDGF-BB-inducible Egr-1 mRNA exposed to MMP-inhibitors 196 5.3.6 PDGF-BB-inducible Egr-1 expression is mediated by EGFR 199 5.3.7 PDGF-BB-inducible Egr-1 expression does not require ErbB2 205

5.3.8 PDGFRβ is not required for EGFR-ligands-induction of Egr-1 expression 208 5.3.9 Inhibition of endogenous HB-EGF, AT II receptors or FGFR-1 do not perturb PDGF-BB-inducible Egr-1 expression 211 5.3.10 PDGF-BB induces PDGFRβ-EGFR complex formation in a MMP/ADAM-dependent manner 215 5.3.11 PDGF-BB-inducible WKY12-22 SMC migration is both EGFR- and MMP-dependent 220 5.3.12 The effect of MMP/ADAM inhibitors on PDGF-BB-inducible Egr-1 expression is cell type-dependent 223

5.4 Discussion 226

6 Conclusion and Future Directions: “Egr-1 Induction by PDGF-BB Involves MMP/ADAM-dependent ERK1/2, PDGFRβ and EGFR Phosphorylation and PDGFRβ/EGFR Heterodimerisation” 232 6.1 Conclusion 233

6.1.1 PDGF-BB-inducible ERK1/2 activation and Egr-1 expression is MMP/ADAM dependent 233 6.1.2 PDGF-BB-inducible Egr-1 expression in neonatal WKY12-22 requires EGFR 236 6.1.3 PDGF-BB promotes cross-talk between PDGFRβ and EGFR in neonatal WKY12-22 SMCs 239 6.1.4 Inhibitors to MMP/ADAM and EGFR attenuate PDGF-BB-induced migration and proliferation in WKY12-22 SMCs 240 6.1.5 Egr-1 induction by PDGF-BB is MMP/ADAM and SMC type-dependent 241 6.2 Future directions 244

6.2.1 Is PDGF-BB-induced c-fos expression MMP/ADAM- and EGFR-dependent? 244 6.2.2 Which MMP(s)/ADAM(s) are controlling PDGF-BB-inducible Egr-1expression? 245 6.2.3 Do different SMCs have dissimilar expression pattern of MMPs, ADAMs, EGFR and PDGFRβ ? 245 6.2.4 Can PDGFRβ heterodimerise with other EGFR isoforms ? 246 6.2.5 Does PDGF-BB stimulation induce EGFR ligand (s) release? 247 6.2.6 Does PDGF-BB activate MMP activity? 247 6.2.7 Do MMP/ADAM inhibitors block EGFR phopshorylation at Tyr845? 248

6.2.8 Can Förster Resonance Energy Transfer (FRET)experiments demonstrate the physical interaction between PDGFRβ/EGFR upon PDGF-BB stimulation and confirm a role of MMP/ADAM in this complex formation? 249

6.2.9 Do second messengers play a role in PDGF-BB-induced EGFR transactivation? 250 6.2.10 Could Egr-1 overexpression overcome the suppressive effect of MMP/ADAM inhibitors on SMC migration and proliferation upon PDGF-BB stimulation? 250 6.2.11 Can EGFR ligands induce PDGFRβ/EGFR assembly? 251

6.2.12 Is ADAM/MMP-dependent Egr-1 expression functional in other cell types, or is it confined to ‘synthetic” SMCs? 251

6.2.13 To demonstrate that Egr-1 induction is both ADAM/MMP in animal models 252

7 Final conclusion of this thesis 253

8 References 256

List of Figures Chapter 1 Figure 1.1 Normal vessel structure 4 Figure 1.2 Schematic representation of atherogenesis 6 Figure 1.3 Smooth muscle cell phenotypes 11 Figure 1.4 Structure of human PDGF/VEGF growth factors domains 13 Figure 1.5 Proteolytic cleavage and activation of the novel PDGF ligands 18 Figure 1.6 PDGF ligands binding to the PDGF receptors 23 Figure 1.7 Signalling pathways for the PDGFRβ 27 Figure 1.8 MAPK signaling cascade 31 Figure 1.9 Schematic representation of the divergent roles of the ATII receptors 34 Figure 1.10 Atherogenic effects of ATII 37 Figure 1.11 MMP structure 39 Figure 1.12 Activation of MMPs 42 Figure 1.13 The chemical structure of GM6001, BiPS and TAPI-1 49 Figure 1.14 Domain structures of the ADAM, SVMP, ADAMTS, MTMMP and MMP metalloenzyme family 51 Figure 1.15 Structural representation of human ADAM-17/TNF-α converting enzyme (TACE) 52 Figure 1.16 Schematic representation of the EGFR 57 Figure 1.17 Schematic representation of the EGFR family binding specificities for EGFR ligands 60 Figure 1.18 Schematic representation depicting the proposed TMPS 66 Figure 1.19 Control elements of the human Egr-1 promoter 71 Figure 1.20 Schematic representation depicting the importance of Egr-1 in various cardiovascular pathological processes 75 Figure 1.21 Schematic representation of structural characteristics of Sp1 79 Chapter 3 Figure 3.1 ATII induces PDGF-C mRNA expression in neonatal SMCs 114 Figure 3.2 ATII induces PDGF-C promoter activation in neonatal SMCs 116 Figure 3.3 ATII does not induce binding to the Egr-1/Sp1 binding site (-35 to -1) on to PDGF-C promoter despite up-regulating Egr-1 mRNA and protein levels 120 Figure 3.4 Proximal region of the PDGF-C promoter 123 Figure 3.5 ATII induces Egr-1 binding to the PDGF-C promoter 126 Figure 3.6 ATII induces Egr-1 binding to the PDGF-C promoter by ChIP 127 Figure 3.7 Egr-1 and Sp1 interact and bind to the PDGF-C promoter 129 Figure 3.8 Egr-1 and Sp1 cooperatively induce PDGF-C promoter activity 131 Figure 3.9 Novel Egr-1 response element in the PDGF-C promoter is responsible for ATII-inducible PDGF-C expression 133 Figure 3.10 Egr-1 DNAzyme blocks ATII-inducible PDGF-C expression 135 Figure 3.11 Egr-1 DNAzyme blocks ATII-inducible Egr-1 expression 136 Figure 3.12 Egr-1 positively influences ATII-inducible activation of the PDGF-C promoter 138 Figure 3.13 ATII-inducible PDGF-C expression acts through the AT1R 140 Figure 3.14 Mutation of the -543/-535 motif represses Egr-1 activation of the PDGF-C promoter 142 Figure 3.15 ATII does not influence PDGF-C mRNA levels in rat adult aortic SMCs despite of inducing Egr-1 mRNA expression 144

Chapter 4 Figure 4.1 Angiotensin II induction of PDGF-C expression is mediated by AT1 receptor-dependent Egr-1 trans-activation 156 Figure 4.2 Proximal region of the PDGF-C promoter 163

Chapter 5 Figure 5.1 PDGF-BB transiently and rapidly induces Egr-1 expression 177 Figure 5.2 PDGF-BB induction of Egr-1 occurs through ERK1/2 178 Figure 5.3 GM6001 inhibits PDGF-BB-inducible Egr-1 mRNA and protein expression in SMCs 181 Figure 5.4 BiPS inhibits PDGF-BB-inducible Egr-1 mRNA and protein expression in SMCs 183 Figure 5.5 Effect of TAPI-1 on PDGF-BB-inducible Egr-1 expression 185 Figure 5.6 Effect of MMP inhibitors on MMP/ADAM mRNA 188 Figure 5.7 Effect of inhibitors to MMP-2, MMP-3 and MMP-9 on Egr-1 mRNA expression levels 191 Figure 5.8 APMA potentiates PDGF-BB-inducible Egr-1 expression and ERK1/2 phosphorylation 195 Figure 5.9 APMA rescues Egr-1 mRNA expression 198 Figure 5.10 Induction of Egr-1 by PDGF-BB is EGFR-dependent 204 Figure 5.11 Role of ErbB2 on PDGF-BB-inducible Egr-1 expression 207 Figure 5.12 EGFR/PDGFRβ cross talk is unidirectional 210 Figure 5.13 PDGF-BB-inducible Egr-1 expression is ATR/FGFR- independent 214 Figure 5.14 PDGF-BB trans-activates EGFR, and promotes association between EGFR/PDGFRβ in a time- and MMP/ADAM-dependent manner 219 Figure 5.15 MMP/ADAM and EGFR inhibitors block PDGF-BB-induced SMCs migration and proliferation 222 Figure 5.16 Effect of MMP inhibitors on Egr-1 mRNA levels in other SMC subtypes 225

Chapter 6 Figure 6 Egr-1 induction by PDGF-BB involves MMP/ADAM-dependent ERK1/2, PDGFRβ, EGFR phosphorylation and EGFR/PDGFRβ heterodimerisation 243

List of Tables Chapter 1 Table 1.1 Expression patterns of PDGF ligands in vascular cells 12 Table 1.2 EGFR subtypes and their corresponding ligands 61 Table 1.3 ADAM involvement in GPCR-EGFR cross talk 65

Chapter 3 Table 3.1 Semi-quantitative PCR conditions and primer sequences used in mRNA gene expression analysis 99 Table 3.2 Real-time PCR conditions and primers sequences used in mRNA gene expression analysis 100 Table 3.3 List of primary antibodies used in Western blotting experiments 107 Table 3.4 ChIP PCR conditions and primer sequences used for amplification of the PDGF-C promoter containing the putative Egr-1 binding site 110

Chapter 5 Table 5.1 List of inhibitors used 169 Table 5.2 Real-time PCR conditions and primers sequences 172 Table 5.3 Antibodies used in Western blotting and Co-IP experiments 174 1

1 Introduction

Cardiovascular disease (CVD) leading to stroke and coronary heart disease, is the principal cause of mortality and morbidity affecting humankind (Gibbons et al. 2008). CVD accounts for approximately one-third of all deaths worldwide (17 million in 1999 and estimated to 25 million in 2020) (NHF 2005). In Australia, in 2002, 37.6 % of all deaths correlated with heart and vascular disease (NHF 2005). In the same year, approximately

3.67 million Australians had coronary heart disease (NHF 2005). This is a major economic burden given the Australian health care system costs on average AU$5.4 billion annually

(AIH&W 2005). Thus, understanding the pathophysiology of cardiovascular events such as myocardial infarction, sudden death, stroke and atherosclerosis remains imperative for scientists and healthcare professionals. Further research into the disease process will potentially prevent the development and progression of cardiovascular disease.

1.1 Atherosclerosis

The process of atherosclerosis is thought to be a chronic immunoinflammatory disease of medium- and large-sized arteries. The onset of the disease may be as early as childhood, or adolescence, and might remain clinically silent until plaque rupture contributes to sudden thrombosis, triggering acute clinical events (Shah 2007). The progression of atherosclerosis involves the gradual build-up of biological material (cholesterol, fibrin, calcium and cellular products) into lesions that could limit blood flow in the vasculature. It is accepted that atherosclerosis is a result of “insults” to the endothelium and vascular 2

smooth muscle cells (SMCs) within the artery wall, some of these include mechanical injury (shear stress and hypertension), viruses or bacteria, oxidised low-density lipoproteins

(LDL) or chemicals from smoking (nicotine) (Ross 1986). Injury to the artery wall promotes the release of growth factors and cytokines, which ultimately initiate immunoinflammatory signalling events and gene activation. These insults activate fibroproliferative and inflammatory responses leading to progression of disease (Ross

1993). It is now considered that cells within the arterial wall (endothelial cells (ECs) and

SMCs together with cells involved in the immune response (macrophages, lymphocytes and neutrophils) are the cellular components responsible for the development of atherosclerotic lesions (Ross 1993).

1.2 Atherogenesis

1.2.1 The artery wall

The artery wall consists of three layers: the intima, the media and the outer adventia

(Figure 1.1). The inner layer of the artery is called the intima. It is composed of connective tissue lined with a compact monolayer of ECs. The single endothelial cell layer is important in maintaining homeostasis and regulating vascular tone. The structural and functional integrity of the endothelium is important for coagulation, immune response reactions and production of the extracellular matrix (ECM)(Newby 2005). The middle layer is known as the media. It is composed of SMCs together with a matrix of collagen, elastin and proteoglycans. An elastic laminae separate the media and the intima. Lastly, 3

the adventitia is the outer layer of the arterial wall, which consists of fibroblasts, collagen, nerves and lymphatics (Adams 2002).

1.2.2 Initiation and progression of atherogenesis

Over thirty years ago, the “response to injury” hypothesis was first postulated by Ross and

Glomset (Ross and Glomset 1973). It proposes that injury to the endothelium will trigger an inflammatory response. Following intimal injury, a series of steps begin to occur as part of this immune response. The damaged endothelium allows increased infiltration of low- density lipoprotein (LDL) into the intima where it can be modified. Modified LDL acts on monocytes and lymphocytes, allowing them to adhere to the endothelium and invaginate through endothelial cell junctions to the intima (Ross 1993). Here, monocytes differentiate to macrophages and upon scavenging modified LDL within the vessel wall, convert into foam cells. These foam cells secrete a number of growth factors and chemokines, which ultimately contribute to the continuous development of plaque. Importantly, foam cells promote the expression of adhesion molecules leading to an ongoing infiltration of inflammatory cells into the artery wall (Li and Glass 2002). In addition, foam cells promote SMC proliferation and migration to the intima at the site of injury (Ross 1986).

Thus, the accumulation of foam cells together with lymphocytes and SMCs result in “fatty streaks”, the first clinically visible change in the arterial wall (Ross 1999). Progression of atherosclerosis is determined by accumulation of SMCs and lipid-laden macrophages in the intima (Figure 1.2).

Figure 1.1 Normal vessel structure. The schematic represents the three compartments of the normal vessel wall; the intima, the media and the adventitia together with the different types of extracellular matrix, basement membranes, interstitial matrix and elastic laminae.

SMC= smooth muscle cells. Adapted from Newby et al. (Newby 2005).

1.2.3 Initiation and progression of atherogenesis: The hypothesis

Over thirty years ago, the “response to injury” hypothesis was first postulated by Ross and

Glomset (Ross and Glomset 1973). It proposes that injury to the endothelium will trigger an inflammatory response. Following intimal injury, a series of steps begin to occur as part of this immune response. The damaged endothelium allows increased infiltration of low density lipoprotein (LDL) into the intima where it can be modified. Modified LDL acts on monocytes and lymphocytes, allowing them to adhere to the endothelium and invaginate through endothelial cell junctions to the intima (Ross 1993). Here, monocytes differentiate to macrophages and upon scavenging modified LDL within the vessel wall, convert into foam cells. These foam cells secrete a number of growth factors and chemokines, which ultimately contribute to the continuous development of plaque. Importantly, foam cells promote the expression of adhesion molecules leading to an ongoing infiltration of inflammatory cells into the artery wall (Li and Glass 2002). In addition, foam cells promote SMC proliferation and migration to the intima at the site of injury (Ross 1986).

Figure 1.2 Schematic representation of atherogenesis. (A) Response to injury. PDGF secreted by platelets is a chemoattractant for SMCs, leading phenotypically modified SMCs

(orange) to migrate from the media into the intima where they proliferate. MMPs also facilitate SMC migration and proliferation by remodelling of the extracellular matrix.

(B) Response to inflammation. PDGF, activated SMCs, ECs (blue) and macrophage foam cells (FCs) promote SMC migration and proliferation. The production of MMPs is stimulated by inflammatory mediators and cell to cell contact through the CD40/CD40 ligand system. Adapted from Newby et al. (Newby 2005). 7

Furthermore, proliferation of SMCs and their production of collagen results in the formation of a fibrous cap. This consists of SMCs, macrophages and extracellular matrix that enclose a lipid rich core. The fibrous cap, inflammatory cells, deposited lipids and necrotic material constitute the atherosclerotic plaque (Libby and Aiwaka 2002).

Alternatively, the fibrous cap can undergo erosion to rupture, exposing circulating blood to the plaque components. As the plaque core is highly thrombogenic, this contact stimulates thrombus formation in the arterial lumen, leading to compromised blood flow or a complete blockage, therefore causing ischaemic events such as myocardial infarction or stroke.

1.2.3.1 Smooth muscle cells (SMCs)

The principal feature of atherosclerotic lesions is the proliferation and migration of SMCs in the intima (Irvine et al. 2000). Proliferation of SMCs is a key feature of atherosclerosis

(Olson et al. 1992), wound repair (Ross 1986), neoplasia (Ross 1986) and growth development . However, SMCs within adult blood vessels proliferate at an extremely low rate. SMCs are highly specialised cells that play a pivotal role in contraction and regulation of blood vessel tone, blood pressure and blood flow distribution (Owens et al. 2004).

Importantly, SMCs can undergo phenotypic change, altering normal structure and function dependent on local environmental cues such as mechanical influences, reactive oxygen species (ROS), cell-cell contact, cell-matrix interactions, increased lipids, exposure to circulating blood products, growth factors and cytokines. This “phenotypic switching” is characterised by increased proliferation and migration, amplified synthesis of ECM and decreased expression of SMC-selective markers. SMC phenotypic switching is responsible 8

for the initiation and progression of diseases including atherosclerosis, cancer and hypertension.

It is now recognised that SMCs from the media and the intima of normal and diseased vessels are heterogeneous (Shanahan and Weissberg 1999). SMCs from the media are spindle-shaped (Lemire et al. 1994), and express a unique repertoire of contractile proteins

(e.g. smooth muscle (SM) α-actin, SM-myosin heavy chain, SM22α, smoothelin, caldesmon) which respond to vasoconstricting and vasodilating agents (Ross 1995), hence these cells are described as “contractile”. On the other hand, SMCs from the neointima of damaged arteries are immature, cobblestone-shaped, express fewer contractile proteins, and significantly express matrix proteins. SMCs of this character have a “synthetic” phenotype

(Campbell and Campbell 1990; Owens et al. 2004), and are a feature of developing and diseased arteries (Figure 1.3). As a result, synthetic cells express and respond to growth factors that induce their migration and proliferation. Interestingly, these cells are more susceptible to apoptosis (Li and Miller 2000). SMC phenotype switching is the consequence of very complex interactions, likely to involve many factors from the surrounding environment. Currently, little is known about how this process occurs in vivo.

1.2.3.2 Environmental cues

Growth factors (such as platelet derived growth factor (PDGF) (Ross 1986), transforming growth factor β (TGFβ) (Seifert et al. 1994), fibroblast growth factor (FGF-2) (Santiago et al. 1999)), cytokines (e.g. interleukin-1 (IL-1) (Nomoto et al. 1988)), and peptide hormones 9

(e.g. angiotensin II (ATII)), regulate SMC proliferation. These molecules are poorly expressed in the un-injured vessel wall, but are secreted during the atherosclerotic process

(Ross 1993). The synthesis and secretion of these factors initiates a signalling pathway which ultimately serves as a chemoattractant signal, directing SMC migration from the media to the intima. In addition to producing these chemoattractant signals, growth factors

(such as PDGF) (Galis and Khatri 2002) are capable of stimulating the production of matrix metalloproteinases (MMPs) by SMCs and macrophages. MMPs degrade and remodel the plaque ECM in which these cells reside. The degradation of the ECM can also induce switching of the contractile SMCs to the migratory phenotype (Cho et al. 2000). A detailed description of MMP synthesis and function can be found in Section 1.5.

1.3 Platelet-derived growth factor (PDGF) family

1.3.1 Overview

Platelet-derived growth factors (PDGFs) were first identified when it was observed that serum, and more importantly platelets, promoted the growth of fibroblasts and SMCs

(Heldin and Wang 1999). PDGFs are a major stimulant of migration and differentiation for a variety of cells of mesenchymal origin including fibroblasts, SMCs and capillary endothelial cells (Utela et al. 2001). PDGFs can also act on cells of non-mesenchymal origin, such as neurons (Utela et al. 2001). However, it is now known that PDGF expression can be induced in all cell types that comprise the intact artery wall, and also in inflammatory cells that ‘infiltrate” the artery in response to pathological stimuli (Table 10

1.1). PDGFs have been implicated in the pathology of different conditions including atherosclerosis, tissue fibrosis and cancer (Fang et al. 2004).

1.3.2 PDGF family of growth factors

PDGFs are members of the PDGF/VEGF (vascular endothelial growth factor) family of growth factors, which consists of VEGF (A, B, C, D, and E), PDGF (A, B, C and D) and placenta growth factors (PlGF) 1 and 2. Structural features common to these growth factors include a high beta sheet content, and a cysteine knot domain that contains 8 cysteine residues perfectly conserved (Figure 1.4). These two characteristics confer high conformational stability of the protein (Utela et al. 2001; Reigstad et al. 2005).

The PDGF ligand chains show high sequence homology, but differ in their chromosomal localisation. The chains form disulphide-linked homo-dimers, PDGF-AA, -BB, -CC, and -

DD and heterodimers PDGF-AB. Notably, the four PDGF chains are inactive in their monomeric forms (Utela et al. 2001). PDGF-A and PDGF-B share 50 % homology, which is comparable to the similarity between PDGF-C and PDGF-D (Heldin and Wang 1999;

Utela et al. 2001; Fang et al. 2004; Reigstad et al. 2005). The five PDGF dimer isoforms exert their effects by binding and activating two different tyrosine kinase receptors, platelet-derived growth factor receptor alpha (PDGFRα) and platelet-derived growth factor receptor beta (PDGFRβ), which bind to the PDGF chains with differential specificity.

Figure 1.3 Smooth muscle cell phenotypes. The differentiation state of SMCs is highly plastic and depends on the integration of diverse environmental cues. Importantly, the phenotypic states that can be displayed by SMCs also depend on the variable expression of

SMC-selective differentiation markers. The presented drawing shows the phenotypes ranging from the highly synthetic/proliferative SMC depicted on the left to the highly contractile, fully differentiated/mature SMC shown on the right. Figure modified from

Owens et al. (Owens et al. 2004).

Table 1.1 Expression patterns of PDGF ligands in vascular cells

Cell Type PDGF-A PDGF-B PDGF-C PDGF-D Reference

Vascular Endothelial + + + - (Raines Cells 2004)

Vascular Smooth + + + +/- [30] muscle cells

Macrophages + + + - (Raines

2004)

Fibroblasts + + +/- + (Raines

2004)

Platelets + + + ? (Raines

2004)

Figure 1.4 Structure of human PDGF/VEGF growth factors domains. The eight invariant cysteine residues are marked by red diamonds. Other invariant amino acids are indicated in blue. Figure taken from Li et al. (Li and Eriksson 2003).

1.3.2.1 PDGF-A and PDGF-B: the classical PDGFs

PDGF-A and -B chains are synthesised as inactive precursor molecules in the endoplasmic reticulum where they dimerise. The dimerised chains are shuttled to the Golgi apparatus for proteolytic removal of the N-terminal pro-peptide before being secreted from the cell

(Heldin and Wang 1999). Hence, they are secreted into the extracellular space in their active forms. Human PDGF-A and -B are located on chromosome 7q22 (Dalla-Favera and

Gallo 1982) and 22q11 (Swan and McBride 1982), respectively.

Knockout studies have demonstrated that both PDGF-A and -B are vital for vascular development, as PDGF-A null embryos die at the embryonic stage whereas PDGF-B knockout mice die just after birth (Betsholtz et al. 2004).

Due to the locality of PDGFRs, PDGF isoforms have been suggested to have paracrine roles in the development of diverse types of mesenchymal cells from different organs.

PDGF isoforms are often expressed by epithelial cells, and PDGF receptors are present on local mesenchymal cells (Betsholtz et al. 2004). The sequence of human PDGF-B is surprisingly similar to the transforming protein of the simian sarcoma virus (SSV) oncogene, v-sis (Doolittle and Hunkapiller 1983), and also bears similarity to its cellular counterpart, the proto-oncogene c-sis (Chiu and Reddy 1984). This homology identifies

PDGF-B as one of the earliest oncogenes, connecting the PDGF/PDGFR signalling to cellular transformation (Reigstad et al. 2005). PDGF-AB and -BB expressed by SMCs and macrophages stimulate proliferation and migration of SMCs through PDGFRβ, while 15

PDGF-AA stimulates protein synthesis and matrix deposition by binding to PDGFRα

(Eriksson et al. 1992). PDGF-AA is a potent mitogen for cardiac fibroblasts, suggesting a role for its receptor, PDGFRα in proliferation (Heldin and Wang 1999).

1.3.2.1.1 PDGF-B

PDGF-B was the first gene of the family to be identified. Advanced human carotid lesions demonstrated an increased level of PDGF-B mRNA compared to normal arteries as assessed by northern blot analysis (Barret and Bemditt 1987). Interestingly, PDGF-B induction was correlated to expression of a macrophage marker, colony-stimulating factor-

1 (csfr-1) or (c-fms), suggesting the association of macrophages and PDGF-B in these lesions (Barret et al. 1984). Intimal SMCs strongly express PDGF-A, and to a lesser extent

PDGF-B. SMC expression of PDGF-B was weaker compared to the expression of this gene in macrophages (Betsholtz et al. 2004). Both PDGF-A and -B were significantly expressed in human tissues at the site of transluminal coronary angioplasty (Ueda et al.

1996). Moreover, in vivo studies of experimentally induced shear stress (Majesky et al.

1990), demonstrated that PDGF-A and -B are up-regulated in vessels with reduced blood flow, as compared to vessels exposed to increased flow. The reduced flow stimulates EC proliferation, thus PDGF-A and -B are expressed at sites of high risk for lesion development. Likewise, in studies involving blood samples from hypercholesterolemic patients (Billet et al. 1996), PDGF-A and -B mRNA were found to be increased 15 to 20- fold in circulating blood mononuclear cells, compared to normocholesterolemic controls, suggesting a relationship between elevated cholesterol levels and increased PDGF-A and -

B mRNA expression in mononuclear cells. Furthermore, in vivo studies demonstrated that 16

infusion of PDGF-BB into rat carotid arteries after gentle filament-mediated endothelial denudation, increased intimal lesion area 15-fold compared to vehicle-treated animals. The bulk of the intimal lesion consisted of SMCs, thus the major in vivo role of PDGF-BB is the mitogenic effect on SMCs from the media into the intima (Jawien et al. 1992). This was also evident in studies performed by Robson et al, where topical application of PDGF-BB in the wounded area stimulated mitogenesis and chemotaxis of fibroblasts, SMCs, and the proliferation of neutrophils and macrophages. (Robson et al. 1992).

1.3.2.2 PDGF-C and PDGF-D: the new members

PDGF-C and PDGF-D contain a unique N-terminal CUB domain (named after the proteins it was first identified in: C for complement, U for sea urchin EGF-like protein and B for bone morphogenic protein-1). Unlike, PDGF-A and -B, PDGF-C and -D are secreted as latent factors, requiring activation by proteolysis to release the growth factor domain from the CUB domain (Fredrikson et al. 2004). The CUB domain plays an important role in binding to the cellular matrix, protein-protein and protein-carbohydrate interactions, as well as regulating the amount of PDGF-C and PDGF-D available to bind and activate the receptor, therefore maintaining PDGF-C and PDGF-D latency (Li et al. 2000). It has been shown that PDGF-CC and -DD are activated by tissue plasminogen activator and urokinase plasminogen activator, respectively (Fredrikson et al. 2004; Ustach et al. 2004). PDGF-C and PDGF-D appear to be expressed at higher levels in normal vessels than PDGF-A, with

PDGF-C predominantly in medial SMCs, and PDGF-D in fibroblastic adventitial cells

(Utela et al. 2001). The fact that PDGF-C and PDGF-D are secreted as latent factors requiring limited proteolytic activation may explain their presence at higher levels in 17

normal vessels over PDGF-A and PDGF-B, which are secreted as active factors (Heldin et al. 2002).

1.3.2.2.1 PDGF-C

PDGF-C gene is located on chromosome 4q32. PDGF-C is a 345 amino acid protein with a two-domain structure: an N-terminal CUB-domain and a C- terminal growth factor domain

(GFD) that shows homology with PDGFs and VEGFs. A hinge region connects the two domains and contains cleavage sites for proteolytic removal of the CUB domains, thereby activating the GFD (Figure 1.5). Cleavage can occur by plasmin and tissue plasminogen activator (tPA) (Reigstad et al. 2003)

PDGF-C mRNA is expressed in numerous human adult tissues, with highest levels in heart, liver, kidney, ovary and pancreas. Smaller amounts are seen in placenta, skeletal muscle and prostate (Li et al. 2000). PDGF-CC is expressed in platelets as well as in vascular

SMCs (Utela et al. 2001; Ustach et al. 2004).

PDGF-C appears to be involved in the three phases of wound healing (Reigstad et al. 2003;

Fears et al. 2005): (i) inflammation, where PDGF-C is highly expressed in alpha granules of platelets, (ii) proliferation, as PDGF-C has been shown to revascularise ischemic mouse heart, and (iii) remodelling, whereby PDGF-C mediates increased levels of metalloproteinase-1 (MMP-1) and its inhibitor. Due to the high expression of PDGF-C in 18

Hinge region

Complement Clr/Cls, sea Urchin, GFD Bone morphogenic protein

Figure.1.5 Proteolytic cleavage and activation of the novel PDGF ligands. PDGF-C and PDGF-D are secreted as latent factors requiring activation in the hinge region with a release of the growth factor domain (GFD) (yellow) from the CUB domain (red). Adapted from Li et al. (Li and Eriksson 2003).

tissues such as placenta, ovary and embryo, PDGF-C has been defined as a potent angiogenic factor (Fang et al. 2004). Li et al. have demonstrated that infusion of PDGF-

CC increased neovascularisation and blood flow in ischemia models (Li et al. 2005).

Interestingly, PDGF-C promoted angiogenesis by triggering VEGF.

Knockout of the PDGF-C gene resulted in perinatal death due to feeding and respiratory complications in mice (Eitner et al. 2002).Since both PDGF-AB and PDGF-C activate

PDGFR (-αβ and -αα), PDGF-C may also be regarded as a powerful mitogen (Eitner et al.

2002; Fang et al. 2004), contributing to tumour formation and fibrotic disease (Reigstad et al. 2003; Fredrikson et al. 2004). The process of fibrosis involves mesenchymal cell proliferation, which can be induced by PDGF stimulation, and tumour growth factor- induced collagen production and deposition (Bonner 2004). Zhuo et al. reported high

PDGF-C mRNA levels following bleomycin-induced lung injury in mice (Zhuo et al.

2004). Further, Grun et al. demonstrated increased PDGF-C in virus-infected heart tissue in mice, where levels of PDGF-C correlated with severity of the inflammatory response in chronic myocarditis (Grun et al. 2005).

An association between PDGF-C and oncogenicity may be observed in the acquired malignant phenotype of NIH/3T3 fibroblasts which overexpress PDGF-C (Ostman 2004).

Cells derived from Ewing sarcoma (a rare tumour which predominantly attacks bone and soft tissue in the young) have been found to express PDGF-C and importantly, dominant negative forms of PDGF-C attenuate the severity of the tumour (Zwerner and May 2001). 20

1.3.2.2.1.1 PDGF-C and atherosclerosis

PDGF-C is normally expressed in the human heart and may play an important role in physiological and pathophysiological cardiac conditions (Pontem et al. 2003). PDGF-C is a ligand for PDGFRα, and its role in inducing proliferation of cardiac fibroblasts has been suggested (Eitner et al. 2002). Nevertheless, the possible role of PDGF-C in atherosclerosis has not been fully investigated.

Transgenic overexpression of PDGF-C in mouse heart induces a progressive cardiac hypertrophy and fibrosis due to an expansion of the fibroblasts in the interstitium, suggesting that up-regulation of PDGF-C plays a role in the development of the pathology

(Li et al. 2000; Pontem et al. 2003). Up-regulation of PDGF-C was shown in mouse virus- infected heart tissue, in which the levels of PDGF-C correlated with the inflammatory process of chronic myocarditis, necrosis of cardiomyocytes, and development of progressive fibrosis (Li et al. 2000). The biological activity of PDGF-C has been confirmed in human coronary artery SMCs in which proliferation was stimulated in vitro

(Utela et al. 2001), suggesting that PDGF-C could be involved in intimal SMC accumulation, an important step in atherogenesis. Due to its ability to bind and activate

PDGFRαα and PDGFRαβ, PDGF-C could potentially play a role in atherosclerosis (Utela et al. 2001; Ustach et al. 2004). PDGF-C promotes SMC outgrowth in the aortic ring assay and stimulates vascularization and healing full-thickness skin excision in a diabetic mouse model of delayed wound healing (Gilberston et al. 2001). 21

In addition, PDGF-C is suggested to play a role in atherosclerosis by stimulating MMP activity and monocyte migration (Wagsater et al. 2009). Furthermore, PDGF-C has been shown to be expressed by macrophages, SMCs and EC in human atherosclerotic lesions

(Fang et al. 2004). Recent work by Karvinen et al. has demonstrated conclusively that

PDGF-C is overexpressed in atherosclerotic human arteries (Karvinen et al. 2009).

1.3.2.3 The PDGF receptors

PDGF receptors are transmembrane proteins which belong to the receptor protein-tyrosine kinase (RPTK) family, possessing intrinsic tyrosine kinase activity (Van der Gee and

Hunter 1994). RPTK signalling involves ligand-mediated receptor dimerisation, as a result, the receptor subunit is trans-phosphorylated (Van der Gee and Hunter 1994), activating therefore the catalytic domains of the sub-cellular receptor. This allows for phosphorylation of the cytoplasmic substrates with a subsequent translocation to the nucleus where the phosphorylated tyrosines will act as docking sites for downstream signalling molecules containing SH2 (src Homology 2-domain-containing) domains (Van der Gee and Hunter 1994).

To date, two PDGF receptors have been characterised: the PDGF receptor alpha (PDGFRα) and the PDGF receptor beta (PDGFRβ). Each receptor contains five extracellular immunoglobulin-like domains (to which ligands bind), a single transmembrane domain and an intracellular tyrosine kinase domain (Ostman 2004). PDGFRs are expressed as monomers but dimerise after ligand binding, resulting in three different combinations -αα, - 22

ββ and - αβ. Importantly, cells expressing only PDGFRα can bind to all the ligands except

PDGF-D whereas PDGFRβ-expressing cells can only bind PDGF-B and -D (Raines 2004).

Schematic representation of PDGF isoform (s) ligand binding to PDGFRs is shown in

Figure 1.6.

Dimerisation of PDGFRs is entirely driven by ligand binding. Ligand binding has the role of cross-linking two receptor molecules in order to increase their local concentration on the cell membrane (Yang et al. 2008). Following cross-linking and autophosphorylation,

PDGFRβ is internalised and degraded by lysosomes in a process that has yet to be elucidated. The cascade of events which culminates with receptor internalisation can be summarised as ligand-binding, receptor dimerisation, activation of intrinsic tyrosine kinase activity and autophosphorylation (Habib et al. 1986). However, it has been hypothesised that signal transduction does not occur solely at the plasma membrane, but also in endosomes. This phenomenon is known as the “signalling endosome hypothesis” which also applies to other receptor tyrosine kinases (RTKs) (Pahara et al. 2010).

PDGF-BB preferentially interacts with PDGFRβ but also has low affinity for PDGFRα

(Eriksson et al. 1992). Hence, it can generate homo- and hetero-dimer receptor assemblies.

PDGF-AA is only capable of uniting PDGFRα to create homo-dimers (Fredrikson et al.

2004). PDGF-CC and PDGF-DD bind to PDGFRα and PDGFRβ respectively (Fang et al.

2004). PDGF-AB has been shown to bind to PDGFRα but it can also stimulate heterodimer formation. 23

Figure 1.6 PDGF ligands binding to the PDGF receptors. PDGFRα and PDGFRβ consist of five extra-cellular immunoglobulin-like domains and an intra-cellular split kinase domain which can homo or heterodimerise. Solid arrows indicate the ability of the different isoforms of PDGF to bind and activate the dimeric receptor complexes. The dashed arrows indicate that receptor heterodimers can be activated. Adapted from Li et al.

(Li and Eriksson 2003).

Similar to the ligands, PDGFRs are encoded by separate genes on different chromosomes

(Kawagishi et al. 1995). PDGFR expression patterns vary according to the type. For instance, SMCs and fibroblasts express both PDGFRα and PDGFRβ (Vassbotn et al. 1994), whereas platelets only express PDGFRα (Schollman et al. 1992). Cells that express both

PDGFRα and PDGFRβ, have the ability to respond differently to ligand binding. For instance, PDGF-BB-induced activation of PDGFRβ stimulates SMC chemotaxis, whereas

PDGF-BB binding to PDGFRα inhibits this phenomenon (Krettet et al. 1997). Normal vessel ECs do not have detectable levels of either PDGFRα or PDGFRβ (Lindner and

Reidy 1995). During embryogenesis, PDGF isoforms are often expressed on epithelial cells in multiple organs, while their corresponding receptors are expressed on neighbouring mesenchymal cells, suggesting a paracrine role for PDGF isoforms.

Gene knockout studies of PDGFRs have demonstrated their necessity for growth and development. PDGFRα-null mice show impaired neural-crest and cardiovascular development, leading to death between E8 to E16 (Li et al. 2000). Mice deficient in

PDGFRβ have a similar phenotype to those lacking PDGF-B, primarily affecting the renal, haematological and cardiovascular systems leading to death at ED17 to ED19 (Li et al.

2000). Furthermore, inactivation of PDGF-B chain (Leveen et al. 1994) or the PDGFRβ

(Soriano 1994) produces a phenotype with poor filtration in the glomeruli and a lethal defect of SMCs in the vessel wall which results in compromised vascular integrity and bleeding at time of birth. 25

PDGFRβ expression has been correlated with progression of atherosclerotic lesions.

ApoE-deficient mice fed a high-fat diet and injected with PDGFRβ antibodies for 12-18 weeks showed reduced aortic sinus lesion development by 67 %, associated with a reduction in SMCs in the fibrous plaque, however no effect was observed when PDGFRα antibodies were administered (Sano et al. 2001). In human atherosclerotic lesions, tyrosine phosphorylation of the PDGFRβ was elevated 5-fold as compared to non-atherogenic tissues (Abe et al. 1998).

1.3.2.4 Signalling via the PDGF receptors

In SMCs, PDGFRα and the PDGFRβ homo- and hetero-dimers induce similar but not identical cellular effects. PDGFRα primarily mediates cellular hyperplasia and hypertrophy, whereas PDGFRβ stimulates mitogenesis and migration (Eriksson et al.

1992). The divergent regulatory roles of each receptor are likely to be a consequence of the activation of distinct signalling pathways.

1.3.2.4.1 Receptor kinase activity

A key characteristic of the PDGFRs is the tyrosine kinase present within the intracellular domain (Bonner 2004; Andrae et al. 2008). This tyrosine kinase is capable of phosphorylating the receptors’ tyrosine residues (tyrosine 857 of the human PDGFRβ), a process known as autophosphorylation (Kazlauskas et al. 1991). Phosphatases are able to inhibit PDGFR signalling transduction via removal of the phosphate from the tyrosine residue following autophosphorylation (Sundaresanan 1995). In addition, mutation on Arg- 26

385 or Glu-390 in the fourth Ig-like domain of the extracellular region of the PDGFRβ, results in impaired PDGF-induced cellular responses, suggesting an important role of these amino acids in the activation of PDGFRβ (Yang et al. 2008).

1.3.2.4.2 Activation of signalling pathways

Within minutes of autophosphorylation, many signalling pathways are initiated by the docking of SH2 molecules at specific autophosphorylated residues of the PDGF receptors

(DeMail et al. 1999). Heldin et al. (Heldin et al. 1998) have identified 11 and 2 autophosphorylation sites on PDGFRα and PDGFRβ, respectively. These sites specifically bind different types of SH2-domain-containing molecules, including tyrosine kinases of the src family, phosphatidylinositol-3’-kinase (PI3K), phospholipase C-γ1 (PLC-γ1), the Grb2-

Sos1 (complex that activates Ras and the mitogen-activated protein kinase (MAPK) pathway/extracellular signal-regulated kinase (ERK)), GTPase-activating protein for Ras

(Ras/GAP), and the transcription factors of the STAT family (Figure 1.7).

PI3K has a number of downstream effector molecules and is implicated in actin reorganisation, chemotaxis, cell growth and anti-apoptosis (Heldin and Wang 1999). PLC-

γ 1 acts on the same substrate as PI3K, phosphatidylinositol 4,5-bisphosphate [PI(4,5)P2].

This catalysis forms inositol 1,4,5-trisphosphate (IP3) and diacylglycerol, which mobilise intracellular Ca2+ from internal stores and activate certain members of the PKC family

(Heldin and Wang 1999). 27

Figure 1.7 Signalling pathways from the PDGFRβ. Binding of the PDGF-BB to the

PDGFRβ initiates downstream signalling pathways which induce a wide range of cellular responses from cytoarchitectural remodelling to proliferation. Figure adapted from Heldin et al. (Heldin et al. 2002) and Stice et al. (Stice et al. 1999).

The non-receptor protein tyrosine kinases (TK) Src, are activated by RPTKs including

PDGFRs (DeMail et al. 1999). src plays a major role in the regulation of proliferation, differentiation, migration, adhesion, angiogenesis, invasion and immune function. src kinase possess SH2 and SH3 domains which allows them to interact with structural and signalling proteins (Lieu and Kopetz 2009). The Src kinase family consists of nine cytosolic protein tyrosine kinases. The best characterised member is the 60kDa, ubiquitously expressed, c-src. All src family members share an N-terminal sequence for membrane targeting, SH2 and SH3 domains for protein binding, a kinase domain and a catalytic domain (Haendeler and Berk 2000). Importantly, the phosphorylation of two tyrosine residues of src, appears to be critical for its activation and/or inhibition. Tyr530 phosphorylation abrogates Src kinase activity, whereas Tyr419 activates its kinase activity

(Schwartzberg 1998).

Grb2 is an adaptor molecule with one SH2 domain and two SH3 domains (Seifert et al.

1994). The domains permit binding of Sos, which converts inactive Ras GDP to active Ras

GTP when the SH2 domain of GrB2 is bound to a receptor phosphotyrosine. Activation of

Ras is of major importance for several cellular responses (Stice et al. 1999). Activated Ras binds to the serine/threonine kinase Raf-1 that initiates activation of the mitogen-activated protein (Lee et al.) kinase cascade, a pathway related to cell growth, migration and differentiation (Cahill et al. 1996).

1.3.2.4.2.1 Mitogen-activated protein kinase (MAPK) cascade

Mitogen-activated protein kinases (MAPKs) are serine/threonine protein kinases which activate transcription factors in response to stimuli such as growth factors (PDGF, FGF-2 and epidermal growth factor (EGF)), cell-matrix interactions and inflammatory signals.

The basic and functional foundation of the MAPK cascade comprises a MAPK kinase kinase (MAP3K) that phosphorylates a MAPK kinase (MAP2K), which then phosphorylates MAPK. There are several MAPK classes, the major ones being extracellular signal-regulated kinases (ERKs) 1 and 2 (ERK1/2), c-Jun NH2-termal kinase-

1, 2 and 3 (JNK-1,-2,-3, also known as SAPK), p38 isoforms of α, β, δ and γ; and-3, 4 and

5 (Chen et al. 2001).

Among the MAPK-mediated pathways, the RAS-MEK1/2-ERK1/2 pathway has been the most extensively characterised (Figure 1.8). Activation of the RAS protein leads to recruitment of RAS to the cell membrane where changes in phosphorylation status results in stimulation of its serine/threonine kinase activity. Once activated, RAS triggers sequential phosphorylation and activation of the MEK1/MEK2 dual-specificity protein kinases for ERK (Montagut and Settleman 2009).

Mitogen-induced mitosis via ERK1/2 activation proceeds by a biphasic manner (Millette et al. 2005). The first activation stimulates cells to progress to G1, occurring within 15 minutes of mitogen stimulation. The second phase (a prolonged activity of ERK) is responsible for progression from G1 to S phase. ERK1/2 translocates to the nucleus where 30

it phosphorylates several transcription factors that induce the expression of multiple genes.

One the most important transcription factors whose expression has been linked to ERK1/2, as well as atherosclerosis, is the early-growth response (Egr-1) (Section 1.6.1).

Figure 1.8 MAPK signalling cascade. Adapted from http://www.cellsignal.com/pathways/map-kinase.jsp

1.4 Angiotensin II

The renin-angiotensin system (RAS) plays an important role in the regulation of a variety of physiological processes. The hormone angiotensin II (ATII) is the most important molecule of this system. Angiotensinogen, produced in the liver, is converted by the enzyme, aspartyl protease rennin, into angiotensin I, which in turn, is cleaved by angiotensin converting enzyme (ACE) resulting in the formation of ATII (Masataka and

Fukuda 2010). Acute exposure to ATII plays a role in regulating renal salt/water balance and systemic vasoconstriction, whereas chronic stimulation induces hyperplasia and hypertrophy of SMCs. Long-term exposure to high levels of ATII is linked to cardiac hypertrophy, reduced fibrinolysis and renal fibrosis (Mehta and Griendling 2007).

ATII acts through two different receptors which are G-protein-coupled receptors, namely angiotensin II type 1 receptor (AT1R) and angiotensin II type 2 receptor (AT2R) (Figure

1.9). The AT1R is a 352 amino acids protein containing seven transmembrane-spanning helices (Haendeler and Berk 2000). AT1R has been implicated in the majority of the pathological processes of CVD (Lemarie and Schiffrin 2010). Generally, ATII effects are mediated by AT1R. This ATII receptor is broadly expressed in organs including the liver, brain, lung, kidney, heart and the vasculature. The ATII-AT1R complex activates diverse signalling cascades, whose effects are time-dependent. For instance, the vasoconstriction effect of ATII, via activation of a G-protein-dependent pathway, occurs in seconds, whereas ATII activation of the MAPK cascade requires minutes, even hours, before effects are observed (Khachigian et al. 2000). In ATII-AT1R-induced signal transduction, one of 33

the earliest events to occur is the activation of src kinases. As previously mentioned, src

2+ kinases have been associated with the generation of second messengers IP3 and Ca

(Paxton 1994). Increased intracellular Ca2+ stimulates the contractility of SMCs and hence may increases migration (Touyz et al. 2001).

In contrast, AT2R has been shown to have anti-proliferative and pro-apoptotic effects in

SMCs by suppressing AT1R (Mehta and Griendling 2007). AT2R is highly expressed in foetal tissue with its expression decreasing after birth, thus suggesting that it might play an important role in foetal formation and development (Khachigian et al. 2000). It is expressed in small amount in the kidney, lung and liver (Mehta and Griendling 2007).

Figure 1.9 Schematic representation of the divergent roles of the ATII receptors. ATII signals through two receptor subtypes, angiotensin II type 1 receptor (AT1R) and angiotensin II type 2 receptor (AT2R). AT1R has been implicated in the majority of the pathological processes of CVD whereas AT2R seems to possess beneficial effects. NO =

Nitric Oxide. Adapted from Lemarie et al. (Lemarie and Schiffrin 2010).

1.4.1 ATII and CVD

Due to its capacity to affect a wide number of different cell types, ATII is a decisive molecule in the development of various diseases. Abnormal ATII levels can lead to diverse pathological situations including inflammation, atherosclerosis, thrombosis and fibrosis which might contribute to abnormalities such as myocardial infarction, stroke, diabetic micro-vascular disease, peripheral vascular disease and congestive heart failure (Mehta and

Griendling 2007). Rajagopalan et al. (Rajagopalan et al. 1996) showed that ATII could increase oxidative stress in ECs leading to dysfunction and a deterioration in ECs relaxation. In addition, ATII has been shown to up-regulate the LDL receptor on ECs, a critical event in atherosclerotic lesion progression (Mehta and Griendling 2007). ATII can also affect monocytes, macrophages, SMCs and ECs by up-regulation of NF-kB-activating cell adhesion molecules such as VCAM-1, ICAM-1 and E-selectin and inflammatory cytokines, IL-6 and IL-8 (Ruiz-Ortega et al. 2000).

ATII is an atherogenic factor (Figure 1.10) as it contributes to the deposition of extracellular matrix proteins (ECM), plays an important role in the migration and adhesion of SMCs, promotes LDL oxidation, impairs nitric oxide (NO) synthesis, amongst other effects in the vasculature. ATII has also been linked with the synthesis of collagen, an important ECM component, via AT1R and AT2R activation (Lemarie and Schiffrin 2010).

ATII receptor antagonists have been useful in identifying mechanisms by which ATII exerts its effects on different molecules. Losartan, an AT1R antagonist, has been shown to inhibit ATII-induced PDGF-A expression, demonstrating AT1R dependence (Day et al.

1999). Moreover, Candesartan CV11974, another AT1R antagonist, confirmed the role of

AT1R in ATII-induced PDGF-B mRNA expression in newborn rat SMCs (Deguchi et al.

1999).

The effect of ATII on SMCs could be defined as “age-related” (Khachigian et al. 2000).

For instance, ATII-up-regulated PDGF-B mRNA levels in neonatal rat SMCs and neointimal SMCs, whereas no expression was observed in adults SMCs exposed to ATII

(Deguchi et al. 1999). Strikingly, the activation of PDGF-A and-B chains by ATII is Egr-1 dependent (Day et al. 1999; Deguchi et al. 1999). ATII has been shown to up-regulate Egr-

1 mRNA levels via AT1R in cardiac rat fibroblasts (Iwami et al. 1996) as well as in neonatal cardiac myocytes (Shamin et al. 1999).

Figure 1.10 Atherogenic effects of ATII. ATII promotes oxidation of LDL and activation of macrophages (foam cells) which in turn induces apoptosis of SMCs and proteolysis of collagen by MMPs. ATII impairs NO synthesis and supports ROS production by ECs, causing EC dysfunction. ATII also promotes adhesion and infiltration of macrophages/monocytes by up-regulating adhesion molecules and chemokines. ATII also promotes periadventitial angiogenesis by up-regulating VEGF expression. Adapted from

Masataka et al. (Masataka and Fukuda 2010).

1.5 Matrix metalloproteinases

1.5.1 Overview

In response to injury or in pathological processes such as atherosclerosis, accumulation of cells and ECM within the intimal layer of the vessel wall results in intimal thickening.

Although ECs, macrophages and SMCs contribute to ECM production, SMCs represent the major source of connective tissue in both un-injured and damaged vessels. Degradation of the basement membrane around SMCs by metalloproteinases (MMPs) facilitates contact with the interstitial matrix, which can ultimately promote the “phenotypic switch” of

SMCs. Parallel to this, growth factors are released, new matrix components are exposed, and once again SMCs are phenotypically modulated to promote migration and proliferation to mediate wound repair (Newby 2005). Thus, MMPs can be as important as growth factors with respect to regulating the behaviour of vascular cells.

1.5.2 Characteristics of the MMP family

MMPs are a family of zinc-containing secreted or surface bound peptidases that are able to degrade almost all of the components of the ECM (Newby 2005; Reuben and Cheung

2006). Structurally, MMPs have three domains that are common to almost all the MMPs

(Figure 1.11), the pro-peptide, the catalytic domain and the hemopexin-like C-terminal

(absent in MMP-7 and -26, replaced by an immuglobulin-like domain in MMP-23) that is linked to the catalytic domain via a hinge region (Newby 2005; Kessenbrock et al. 2010).

Figure 1.11 MMP structure. The domain structure common to MMPs is shown.

Variations on this domain and other domains on different MMPs are also described. Tm, trans-membrane domain; P, represents a glycosylphosphatidylinositol (GPI)-anchor membrane. Adapted from Newby et al.(Newby 2005).

MMPs are predominantly expressed in an “inactive” form due to the interaction of the cysteine residue of the pro-domain with the Zn2+ of the catalytic site (Figure 1.12). The pro-domain spans a consensus sequence that requires an intracellular (e.g. furin) or extracellular (e.g. plasmin) proteolytic cleavage by convertases. The subsequent proteolytic removal of the pro-domain, a process known as “cysteine switch”, allows the enzyme to become active (Sternlicht and Werb 2001). Most cells express MMPs however some are often associated with a specific cell type, for example, MMP-2 is usually restricted to connective tissue cells with a rare expression in epithelial cells. Similarly,

MMP-7 expression is associated with glandular epithelial cells of the small intestine, but not present in the stromal cells of this organ (Reuben and Cheung 2006).

Classified according to the substrate they catalyse, MMPs are divided into four sub-groups:

(1) the “collagenases”, including MMP-1, -8, -13, which degrade collagen type I, II and III;

(2) the “gelatinases”, MMP-2 and -9, which degrade collagen type IV in the basement membrane; (3) the “stromelysins”, MMP-3, -10 and -11, which degrade non-collagen matrix proteins such as fibronectin, laminin, aggregan; and a “miscellaneous” group (4) which includes MMP-7 (matrilysin) and MMP-12 (elastase). In addition, some MMPs resemble membrane-type MMPs (MT-MMPs) which are directly attached to transmembrane domains MT1-, MT2-, MT3- and MT5-MMP or by glucophosphatidyl inositol anchors (MT4- and MT6-MMP) (Newby 2006). Although MMP-2 and MMP-9 have similar substrate-specificities, there are differences controlling their expression.

MMP-2 is constitutively expressed by SMCs and its expression is not regulated by growth factors or cytokines (Fabunmi et al. 1996). 41

In contrast, MMP-9 is expressed at a low level basally and its activity is induced by IL-1 and TNF-α but not PDGF (Webb et al. 1997). Interestingly, after vascular injury, MMP-9 mRNA and protein expression are up-regulated whereas MMP-2 expression patterns remain unchanged (Cho et al. 2000).

An important difference between MMPs and MT-MMPs is that MT-MMPs are activated intracellularly. A furin recognition motif localised between the pro-domain and the catalytic domain is cleaved by pro-protein convertases in the trans-Golgi apparatus, ensuring that MT-MMPs are active when they reach the membrane (Sounni and Noel

2005). In vitro activation of MMPs can be achieved by the use of thiol-modifying agents such as 4-aminophenylmercuric acetate (APMA), mercury chloride, N-ethylmaleimide, oxidised glutathione or sodium dodecyl sulphate (Newby 2006). These compounds mainly cause disturbance of the cysteine-zinc interaction at the cysteine switch.

MMPs play a pivotal role in normal physiological processes and their dysregulation can therefore be associated with pathological states. Knock-out studies have validated this notion. MMP-2 knock-out mice suffer alterations in body size (Fingleton 2006; Page-

McCaw et al. 2007); MMP-3 knock- out mice show improper neuromuscular junctions

(Fingleton 2006). MMP-7 knock-out mice show defects in innate immunity (Wilson et al.

1999), MMP-9 knock-out have impaired angiogenesis (Vincenti and Brinckerhoff 2007).

In MMP-9, -13 and -14 knock-out mice bone remodelling abnormalities are observed

(Page-McCaw et al. 2007). 42

Figure 1.12 Activation of MMPs. Diagram showing the cysteine switch mechanism for activation of MMPs. Cysteine in the proenzyme domain is associated with zinc to maintain latency of the enzymes. Sodium dodecyl sulphate (SDS) can unfold the structure to expose zinc. N-ethylmaleimide (Ohtani et al.), glutathione (GSSG), hypoclorous acid (HOCl) and organo mercurials such as APMA inactivate the cysteine. Alternatively, proteolytic enzymes cleave the propeptide. In the second step, active forms can be autocatalytically cleaved by the already activated MMPs to remove the pro-peptide. GM6001 and BiPS inactivate MMPs by chelating Zn2+. Taken from Chakraborti et al.(Chakraborti et al.

2003). 43

1.5.3 Regulation of MMPs

Proteolytic activity of MMPs is tightly regulated at several levels including gene expression, post-translational modifications and chromatin remodelling (Chang and Werb

2001).

1.5.3.1 Gene expression

A new classification of MMPs based on mechanisms regulating their expression has been described (Yan and Boyd 2007). Group I comprises MMPs with a TATA box and an activator protein-1 (AP-1) site located in the proximal region (~-70bp) of their promoters, which binds members of the Jun and Fos family of transcription factors (e.g. MMP-1, -3, -

7, -9, -12, -13,-19 and -26). Group II, (MMP-8, -11 and -18) where a TATA box is present but not an AP-1 site. Group 3 (MMP-2, -14 and -28) have neither a TATA box, nor the

AP-1 site, implying that these MMPs are constitutively expressed.

MMPs are induced by multiple stimuli, including inflammatory cytokines such as interleukin-1 (IL-1), IL-4 and tumor necrosis factor-α (TNF-α), mechanical movement and phagocytosis. The promoters of the MMPs also have binding sites for a variety of transcription factors. Ets has been shown to increase basal activity of the MMP-1 promoter

(Westmarck et al. 1997). Moreover, Sp-1, Sp-3 and AP-2 transcription factors have been shown to activate expression of the MMP-2 gene in astroglioma cells (Qin et al. 1999).

Likewise, Egr-1 is implicated in the regulation of MMP-14 in cells cultured in three- dimensional matrices (Barbolina et al. 2007), and MMP-2 by cigarette smoke in lung 44

fibroblasts (Ning et al. 2007). Signal transduction pathways such as ERK1/2 and the p38 can also induce expression of multiple MMP genes (Chang and Werb 2001; Yan and Boyd

2007). The former is widely linked to activation of MMP-1, -3, -7, -9 and -14 by EGF.

Likewise, IL-1 activates MMP-2 via this pathway. p38 is correlated with activation of

MMP-1, -3, -9 and -13 in response to inflammatory signals (Yan and Boyd 2007).

1.5.3.2 Post-translational modifications

Amongst post-translational modifications, phosphorylation may play an important role in controlling MMP activity. A study by Sariahmetoglu et al. (Sariahmetoglu et al. 2007) demonstrated that when MMP-2 is phosphorylated by protein kinase C (PKC) at residues

S32, S365, T250 and Y271, its activity is largely reduced.

1.5.3.3 Chromatin remodelling

Chromatin remodelling also plays a role in MMP expression. Specifically, hystone acetylation regulates the expression of a number of MMP genes. Acetylation weakens the histone:DNA interaction, allowing access to transcriptions factors and therefore increases gene activation (Clark et al. 2008). Acetylation is a reversible modification with acetyl groups added by a family of histone acetyl transferases (HATs), and removed by histone deacetylases (HDACs). HDAC inhibitors (HDACi) can enhance IL-1 or TNF-α induction of MMP-3, and also repress IL-1 or TNF-α induction of MMP-1 and MMP-9 transcription and translation (Pender et al. 2000). Activated NF-κB increases the expression of several 45

MMP genes in response to proinflammatory stimuli. Activation of NFκB can be modulated by histone acetyl transferases (HATs), including p300 and CBP.

Methylation, another chromatin remodelling process plays a role on controlling MMP expression. Methylation is usually associated with repressive chromatin state and inhibition of gene expression, thus, it may block the binding of transcriptional activators and/or methyl binding proteins may recruit transcriptional repressors including HDACS

(Clark et al. 2008). Methylation of the MMP-9 promoter is inversely associated with its gene suppression in lymphoma cells (Yan and Boyd 2007). Similarly, a colon cancer cell line with a defective methyl-transferases show increased expression of MMP-3 (Couillard et al. 2006). Hypomethylation of CpG sites in the promoters of MMP-3, MMP-9 and

MMP-13 correlates with the increased expression of these MMPs by chondrocytes in osteoarthritis (Roach et al. 2005).

1.5.3.4 Tissue inhibitors of MMP

Considering the extensive roles of MMPs in biological functions, their physiological activity must be tightly regulated in normal vessels. Tissue inhibitors of matrix metalloproteinases (TIMPs) are a family of at least four endogenous proteins (TIMP-1, -2, -

3 and -4) of approximately 23kDa that reversibly inhibit MMPs by binding to their active site in a stoichiometric 1:1 molar ratio, thereby blocking access to their substrates

(Sternlicht and Werb 2001; Vanhoutte and Heymans 2010). Notably, the low level of sequence homology among the TIMPs might indicate unique and distinct biological roles 46

for each TIMP. TIMPs do not necessarily have a preferable substrate however TIMP-2 shows, to a certain degree specificity for MMP-2, and TIMP-1 for MMP-9 (Chow et al.

2007). TIMP-2 suppresses growth-factor responsiveness as it may interfere with the activation of tyrosine kinase type-growth factor receptors (Hoegy et al. 2001). TIMP-3 is different from the other TIMP members, as it can inhibit ADAMs (a disintegrin and metalloproteinase) (Sternlicht and Werb 2001), which can interact with the ECM

(Vanhoutte and Heymans 2010). TIMP-4 is abundantly found in the heart which might suggest a protective effect against oxidative stress injury (Chow et al. 2007; Vanhoutte and

Heymans 2010). An excess of MMPs over TIMPs results in ECM destruction leading to atherosclerosis and vascular remodeling (Galis and Khatri 2002).

TIMPs also have mitogenic effects on several cell types however the mechanisms behind this positive effect remain unclear. TIMP-1 displays this characteristic. It was first described as EPA for its erythroid-potenciating-activity (Gomez et al. 1997). Depletion of

TIMP-1 suppressed erythropoietin-induced differentiation of mouse erythroleukemia cell line, suggesting that EPA action of TIMP-1 on erythroid leukemia cell lines is related to its activity to promote cell growth (Murate et al. 1993).

Other endogenous MMP inhibitors with irreversible effects, e.g. α2-macroglobulins, have also been identified. These MMP inhibitors are found abundantly in plasma, representing the major source of inhibition for MMPs, whereas TIMPs may only act locally (Sternlicht and Werb 2001). α2-macroglobulins repress MMP activity by trapping the MMPs within 47

the macroglobulin after proteolysis (Barret 1981). As MMPs are implicated in neointima formation, TIMPs or other MMP inhibitors might be seen as a potential treatment to prevent neointimal lesions. It has been demonstrated that TIMPs reduce intimal migration of SMCs and TIMP-3 can promote SMC apoptosis (White and Newby 2002).

1.5.3.5 Pharmacological inhibitors of MMPs

A number of pharmacological MMP inhibitors (MMPi) have been used in in vitro studies with great promise. However, it is uncertain how well these results translate to use in vivo as therapeutic agents in the treatment of cancer, inflammation and CVD. Clinical trials of

MMPi have been unsuccessful, most probably due to the broad-spectrum effects of these inhibitors. Basal expression of MMPs is needed for maintaining normal functions. The consequence of blocking basal MMP activity systematically can cause severe toxicity in the musculo-skeletal system (Vincenti and Brinckerhoff 2007). Synthetic MMPi have been designed by exploiting the Zn2+-dependent mechanism by which MMPs cleave their substrates. The majority of the MMPi are formed of two parts, a peptidomimetic backbone, which allows the MMPi to fit in the recognition pocket of the enzyme and a zinc binding group which binds to the zinc ion (Fears et al. 2005). Early MMPi were comprised of carboxylates, thiols, hydroxamic acids and phosphonic acids (Skiles and Gonnella 2001).

Amongst these, hydroxamic-derivatives have emerged as a preferred Zn2+ chelator due to its potent MMP binding and relative ease of synthesis. However, the aforementioned issues with toxicity still apply to this MMPi group. The broad-spectrum inhibitor GM6001, also known as Galardin or Ilomastat (Figure 1.13A, B), has been shown to possess very potent antimetastatic effects in a transgenic mouse model for breast cancer (Almholt et al. 2008). 48

GM6001 can also reduce phorbol ester-induced cutaneous inflammation and hyperplasia

(Hollera et al. 1997), and inhibit ERK activation and EGFR phosphorylations in cells exposed to bombesin or lysophosphatidic acid (LPA) (Santiskulvong and Rozengurt 2003).

(2R)-N-hydroxy-3-phenylpropioanamide, known as MMP-2/MMP-9 inhibitor or BiPS,

(Figure 1.13C) is another example of an MMPi. It is a derivative of sulfonylamine acid, demonstrated to have beneficial effects such as reducing lung colonisation by Lewis lung carcinoma cells in mice (Tamura et al. 1998), blocking the invasion of mouse brain microvessel ECs in Matrigel (Fears et al. 2005) and inhibiting TGF-β-induced cataract formation in the rat lens (Dwivedi et al. 2006). Furthermore, TAPI-1 (Figure 1.13D), the

ADAM-17 inhibitor, has been shown to decrease urokinase-induced migration of human

SMCs by inhibiting EGFR transactivation (Bakken et al. 2009). TAPI-1 has been shown to also inhibit insulin-induced SMCs proliferation by blocking phosphorylation of the EGFR

(Roztocil et al. 2008).

A. GM6001 Inactive analogue B. GM6001

C. BiPS

D. TAPI-1

Figure 1.13 The chemical structures of (A, B) GM6001 (Adapted from Sanstiskulvo et al

. (Santiskulvong and Rozengurt 2003)), (C) BiPS (Adapted from Saito et al. (Saito et al.

2002)), and (D) TAPI-1 (http://www.merck-chemicals.com/usa/life-science-research/tapi-

1/EMD_BIO579051/p_WGCb.s1L7sYAAAEW0mEfVhTm?WFSimpleSearch_NameOrID

=TAPI&BackButtonText=search+results The circled structure represents the hydroxamic group. 50

1.5.4 ADAM family

The ADAMs (a disintegrin and metalloproteinase) family of enzymes are Zn2+ dependent, transmembrane and secreted proteins which are important in cell adhesion and in proteolytic cleavage of the ectodomains of a diverse range of cell surface receptors

(Edwards et al. 2008). ADAMs are closely related to other metalloenzymes such as

MMPs, ADAM-TSs (ADAMs with thrombospondin domains) and SVMP (snake venom metalloproteinase) (Figure 1.14). In the human genome, 21 ADAMs have been identified, but only 13 have been found to be proteolytically active (Edwards et al. 2008). The proteolytically inactive ADAMs (1-7, 22, 23, 29, 31 and 32) are believed to participate in intercellular communication (Gooz 2010).

The structure of the ADAMs consists of conserved protein domains including a NH2- terminal signal sequence, followed by a pro-domain, a metalloproteinase domain, a disintegrin domain, a cysteine rich region, an EGF-like domain, a trans-membrane domain and a cytoplasmic domain (Figure 1.15). The pro-domain keeps the enzyme inactive until is removed by a furin-type pro-protein convertase or autocatalysis. Similar to the role of the cysteine switch in the MMPs structure, the cysteine rich region appears to maintain the metalloproteinase domain in an inactive state (Ohtsu et al. 2006). However, the cysteine switch is not essential for inhibition of the enzymatic activity of ADAM-17, but its pro- domain has inhibitory properties (Gonzales et al. 2004).

Figure 1.14 Domain structures of the ADAM, SVMP, ADAMTS, MTMMP and MMP metalloenzyme family. Adapted from Edwards et al. (Edwards et al. 2008).

Figure 1.15 Structural representation of human ADAM-17/TNF-α converting enzyme

(TACE). A pro-domain contains the cysteine switch box motif (PKVCGY). A catalytic- site consensus motif is located in the metalloproteinase domain (HEGLH). Thr735 and

Ser819 are cytoplasmic phosphorylation (P). PXXP motifs are predicted sites for associating with SH3 domains. Adapted from Ohtsu et al. (Ohtsu et al. 2006).

The ADAM disintegrin-like domain is known to interact with integrins. These interactions have been shown to control cell adhesion and cell-cell interactions (Edwards et al. 2008).

The cytoplasmic domain plays a role in coupling ADAMs, particularly ADAM-17, to specific signalling events such as G protein-coupled receptor (GPCR) activated ADAM- mediated EGFR ligand release (further discussed in Section 1.5.5.3)

The most well known function of ADAMs is cleavage of various transmembrane proteins:

EGFR ligands, pro-inflammatory cytokines, adhesion molecules and amyloid precursor protein. After cleavage, free molecules bind either to their receptors on the same cell, receptors on neighbouring cells, or alternatively they can reach more distant cells in the same tissue or even enter the bloodstream. Subsequent to ligand binding, the activated receptor initiates downstream signalling events (Gooz 2010).

The expression patterns of the ADAMs are quite unique for each particular set, thus indicating tissue-specific function. For example, some ADAMs (ADAM-2, -7, -18, -20, -

21, -29 and -30) are expressed in the testis, suggesting a role in spermatogenesis and sperm function; whereas ADAM-9, -10, -12, -15, -17 and -19 are expressed in somatic tissues

(Edwards et al. 2008). ADAM-11, -22 and -23 expression is considerable in the central and peripheral nervous systems and ADAM-12 in the placenta and mesenchymal cells (Clark et al. 2008).

Like MMPs, ADAMs are also inhibited by TIMPs. In fact, TIMPs appear to be more selective to ADAMs than MMPs. TIMP-3, for example, can functionally inhibit almost all the proteolytically active ADAMs (Edwards et al. 2008). ADAM-10 can be inhibited by

TIMP-1 and TIMP-3, whereas ADAM-8, -9 and -19 do not respond to TIMP inhibition

(Amour et al. 2002). Synthetic inhibitors also demonstrate selectivity for specific ADAMs;

G1254023X inhibits ADAM-10 whereas GW280264X inhibits both ADAM-10 and

ADAM-17.

1.5.4.1 ADAM-17

One of the most well studied ADAMs is ADAM-17. It was discovered in 1997, as the

“enzyme that releases the membrane-bound TNF-α precursor to a soluble form” (Black et al. 1997). ADAM-17, also known as TACE, is widely expressed in various tissues, including the brain, heart, kidney, and skeletal muscle (Black et al. 1997). ADAM-17 has very little sequence homology with other ADAMs, the closest being ADAM-10. There is evidence that ADAM-17 is “kept” in lipid rafts during its transport and maturation through the Golgi apparatus, this process is believed to help maintain its latency keeping ADAM-17 separated from its substrates (Gooz 2010). The catalytic domain containing the Zn2+ is responsible for cleaving membrane bound proteins such as TNF-α. The cytoplasmic domain of ADAM-17 contains serine and threonine residues which require phosphorylation for its activation (Ohtsu et al. 2006). There is evidence that ERK-dependent phosphorylation of Thr735 is necessary for ADAM-17 to reach its secretory pathway during maturation (Soond et al. 2005). 55

ADAM-17 enzymatic activity is increased in diseases in which levels of TNF-α are found to be elevated, such as rheumatoid arthritis and inflammatory bowel disease (Gooz 2010).

However, the role of ADAM-17 in inflammation is not just limited to its TNF-α sheddase activity (ability of MMPs and ADAMs to cleave membrane proteins at the cell surface, releasing soluble ectodomains with altered location and function) (Nabel et al. 1993).

ADAM-17 has been identified as the enzyme cleaving L-selectin, therefore promoting leukocyte migration into the vascular endothelium (Wang et al. 2009). Other studies have recognised ADAM-17 as the sheddase for vascular cell adhesion molecule (VCAM) which mediates leukocyte adhesion to the vascular endothelium (Garton et al. 2003). Studies on

ADAM-17 have also suggested a role in acute myocardial infarction and cardiac remodelling. Elevated ADAM-17 expression has been observed in atherosclerotic plaques of the aortic arch and sinus in apolipoprotein-E-deficient mice and in human lesions

(Canault et al. 2006). Similarly, ADAM-17 and TNF-α expression has been shown to be increased in areas of ruptured coronary plaques in patients with myocardial infarction

(Satoh et al. 2008). Inhibition of ADAM-17 by small interfering RNA has been correlated with inhibition of both ATII-induced and hypertension-induced cardiac hypertrophy and fibrosis (Wang et al. 2009). ADAM-17 has also been implicated in carcinogenesis because it sheds growth factors required for tumour progression, growth and inflammation.

Generally, the association between the development and progression of malignancies with

ADAM-17 is that ADAM-17 is able to cleave different epidermal growth factor (EGF) ligands. The expression of EGFR ligands is in turn correlated with poor disease progression (Gooz 2010).

1.5.5 EGFR

The EGFR also known as ErbB1 or HER1 (Figure 1.16), is a 170 kDa transmembrane glycoprotein characterised by an extracellular ligand-binding domain with two cysteine- rich regions, a single α-helical transmembrane domain and a cytoplasmic domain which contains the tyrosine kinase region (Citri and Yarden 2006). The EGFR family consists of four transmembrane receptors that include ErbB2 (HER2, Neu), ErbB3 (HER3), and ErbB4

(HER4), in addition to the aforementioned EGFR (ErbB1 or HER1) (Berasain et al. 2009).

The tyrosine kinase region is followed by a carboxy-terminal tail which harbors the autophoshorylation sites. Importantly, this domain is well conserved within the EGFR family except in ErbB3 in which some amino acids are changed, resulting in impaired tyrosine kinase activity (Sibilia et al. 2007).

The extracellular ligand-binding domain is less conserved among the EGFR members, most likely due to their ligand binding specificities. To date, no ligands have been identified for

ErbB2 (Sanderson et al. 2006; Higashiyama et al. 2008). Similar to PDGFRs, ligand- binding results in homo- or hetero-dimer EGFR formation and cross-phosphorylation of each other (Citri and Yarden 2006). The autophosphorylation at distinct tyrosine residues creates a docking platform for various signalling proteins such as Shc, Grb7, Grb2, Crk, phospholipase Cγ (PLCγ), Src, PI3K, the phosphatases SHP1 and SHP2, and the Cbl E3 ubiquitin ligase (Schneider and Wolf 2008).

Figure 1.16 Schematic representation of the EGFR. Taken from http://www.cellsignal.com/pdf/2239.pdf

Three major autophosphorylation sites (Tyr 1068, 1148 and 1173) and two minor sites (Tyr

992 and 1086) have been documented for EGFR (Bishayee et al. 1999). These sites are clustered in the last 194 amino acids in the C-terminal domain of the receptor (Bishayee et al. 1999). Transcription factors downstream of these docked protein molecules are then activated, including Sp1, Egr-1 and the proto-oncogenes fos, jun and myc. Activation of these signalling pathways will then affect proliferation, migration, cell adhesion, growth inhibition and differentiation (Sanderson et al. 2006). Significantly, immunohistochemical studies have identified EGFR expression on intimal SMCs within human atherosclerotic plaques (Tamura et al. 2001). EGFR has also been shown to mediate cell proliferation and

DNA synthesis in cultured rat aortic SMCs (Nanney et al. 1988). Furthermore, EGFR has been demonstrated to co-localise with macrophages within atherosclerotic lesions in cholesterol-fed rabbits (Lamb et al. 2004).

1.5.5.1 EGFR-ligands

A diverse range of growth factors can bind EGFR, these include: EGF, TGF-α, amphiregulin (AR), epiregulin (EREG), β-cellulin (BTC), epigen (EPG), a complex subfamily of Neuregulins (NRG1-6) and HB-EGF (Sanderson et al. 2006; Berasain et al.

2009) (Figure 1.17). These EGFR-ligands are expressed as Type 1 transmembrane precursor proteins, which contain unique domains: (i) an EGF-like domain that establishes the receptor-ligand specificity; (ii) an immunoglobulin-like domain; (iii) a hydrophobic transmembrane domain; (iv) a hydrophilic cytoplasmic tail, and (v) other additional domains which include glycosylation and heparin-binding domains in AR and HB-EGF. 59

Heparin-binding domains bind to heparan sulfate chains in cell surface heparan sulfate proteoglycans (HSPGs), which are required by AR and HB-EGF for efficient EGFR activation (Raab and Klagsbrun 1997). As these ligands are membrane-anchored immature proteins, they need to be activated by catalytic removal of their extracellular domain which is typically performed via MMP. This shedding liberates a soluble growth factor which is ready to bind the EGFR (Berasain et al. 2009). The distinctive binding patterns of the

EGFR ligands is summarised in Table 1.3.

1.5.5.1.1 EGFR ligands and atherosclerosis

EGF, HB-EGF, TGF-α, BTC and EPR have been shown to mediate SMC phenotypic switching, essential to the development and progression of atherosclerosis (Yamakana et al.

2001). HB-EGF is a more potent mitogen for SMCs than EGF and TGF-α (Massague and

Pandiella 1993). Interestingly, intimal SMCs appear to be more responsive to the mitogenic effects of EGF compared to medial SMCs (Mitsumata et al. 1994). HB-EGF expression is up-regulated in human atherosclerotic plaques (Reape et al. 1997), as well as in neointimal cells of rat carotid arteries in response to balloon injury (Igura et al. 1996).

Likewise, BTC expression levels is increased in the aortae of individuals with atherosclerosis (Tamura et al. 2001). Similarly, AR (Kato et al. 2003) and EPR (Toyoda et al. 1997) have been shown to stimulate SMCs in vitro and in vivo (Kato et al. 2003)

(Toyoda et al. 1997).

Figure 1.17 Schematic representation of the EGFR family binding specificities for

EGFR ligands. EGFR ligands can be divided in four groups according to their binding preferences. Group 1(EGF, TGFα, AR and Epigen) binds to receptor pairs involving

EGFR (ErbB1) (yellow). Group 2 (BTC, HB-EGF and ER) binds EGFR (ErbB1) and

ErbB4 (blue). The third group (NRG-1 and -2) bind to ErbB3 and ErbB4 (red). The fourth group (NRG-3, -4 and -5) bind to ErbB4 (green). Adapted from Sanderson et al.

(Sanderson et al. 2006).

Table 1.2 EGFR subtypes and their corresponding ligands. Taken from Schneoder et al

(Schneider and Wolf 2008), Massague et al. (Massague and Pandiella 1993)

EGFR/Erb1/HER1 HER2/Erb2 HER3/Erb3 HER4/Er42

EGF Neuregulin 1 HB-EGF

TGFα No identified Neuregulin 2 Betacellulin

Amphiregulin high Epiregulin

HB-EGF affinity Neuregulin 1

Betacellulin ligand Neuregulin 2

Epiregulin Neuregulin 3

Neuregulin 4

1.5.5.2 EGFR activation by MMPs/ADAMs

EGFR-ligands are expressed as transmembrane pro-forms, requiring proteolytic cleavage by activated MMPs to release the soluble and mature forms in a process called “ectodomain shedding”. Following release from the membrane, the ligand is free to bind and transactivate EGFR.

The precise mechanism(s) by which stimuli activate ADAMs with subsequent up- regulation of ectodomain shedding is unknown. Nevertheless, few pathways have been suggested: (i) through the src homology (SH3) domain binding sites contained by ADAM and (ii) via ADAM phosphorylation, which in turn may lead to the binding of intracellular signalling molecules or adaptor proteins (Sanderson et al. 2006). Notably, this process has important biological implications as it places the EGFR signalling system as a convergent point for cell proliferation, survival, and migration and inflammatory responses. The pro- tumorigenic cytokine TGF-β is also associated with expression of EGFR ligands, HB-EGF and TGF-α, through the activation of the transcription factor NF-κB (Murillo et al. 2007).

Furthermore, TGF-β is able to activate ADAM-17 which cleaves EGFR ligands from the plasma membrane allowing EGFR binding (Murillo et al. 2005).

1.5.5.3 G-protein coupled receptors (GPCR)-EGFR cross talk: triple membrane passing

signalling

GPCRs represent the largest group of cell surface receptors (Gutkind 1998; Mifune et al.

2005). They respond to a plethora of stimuli: hormones, neurotransmitters, vasoactive peptides, odorants and light (Gutkind 1998). GPCRs contain seven transmembrane domains and lack enzymatic activity. Intracellularly, a G-protein is coupled to a receptor.

The G-protein consists of 3 subunits; α, β and γ. The G-protein is inactive when the α subunit is bound to GDP (Mifune et al. 2005). Following agonist binding to the GPCR, the

GDP is exchanged for GTP, activating the G-protein. Upon activation, the subunits dissociate from the receptor and the G-protein separates into an α-GTP and a βγ complex

(Prenzel et al. 1999). These dissociated G-protein complexes are free to initiate intracellular activities ,such as activation of ADAMs, cleaveage of EGFR ligands and activation of EGFR (Table 1.3). The first report on this transactivation was presented in

1999 by Prenzel and colleagues (Prenzel et al. 1999). The transactivation mechanism is described as the “triple membrane-passing signalling” and is illustrated in Figure 1.18.

Essentially, it delineates the mechanisms involved in GPCR-induced activation of

MMPs/ADAMs followed by shedding of the EGFR ligand HB-EGF. In this study,

Batimastat, a broad-spectrum MMPi inhibited HB-EGF shedding, confirming a role for

MMP activity in this process (Prenzel et al. 1999). In addition, another antagonist, CRM-

197, a mutant form of diphtheria toxin that blocks HB-EGF function , was shown to inhibit

EGFR transactivation, confirming HB-EGF as the ligand responsible for EGFR transactivation (Prenzel et al. 1999).

ATII signalling through AT1R is also associated with EGFR transactivation, likely to be mediated by ADAM-17-induced HB-EGF shedding, in different cell types, including

SMCs (Eguchi et al. 2001) and cardiomyocytes (Thomas et al. 2002). Interestingly,

ADAM-17 is required for ATII-induction of TGF-α shedding in the kidney (Lautrette et al.

2005). ADAM-12 is activated by ATII to release HB-EGF in cardiomyocytes (Asakura et al. 2002). Lysophosphatidic acid (LPA) binds a subfamily of GPCRs which belong to the

LPA receptors. LPA transactivates the EGFR which in turn activates the ERK signalling pathway culminating in cell growth (Anliker and Chun 2004). Other GPCR agonists implicated in transactivation events are ATP, thrombin, endothelin-1, carbachol, serotonin and membrane type bile acid receptor (Gooz 2010). Different ADAM-mediating EGFR ligand sheddases have been named, e.g. ADAM-10, -12, -15 and -17 (Table 1.3) (Ohtsu et al. 2006). Additionally, some MMPs have also been associated with EGFR transactivation by GPCR (Higashiyama et al. 2008). MMP-2 and -9 have been demonstrated to mediate

HB-EGF shedding through activation of G-coupled gonadotropin-releasing hormone receptor by gonadotropin-releasing hormone in gonadotropic cells (Roelle et al. 2003).

Table 1.3 ADAM involvement in GPCR-EGFR cross talk

Agonist GPCR ADAM EGFR ligand EGFR References Type

Angiotensin II AT1R ADAM- HB-EGF, EGFR (Schafer et 17 TGFα al. 2004; Lautrette et al. 2005)

Lysophosphatidic LPA ADAM- TGFα EGFR/ERbB2 (Schafer et receptor 15 al. 2004; acid (LPA) Schafer et ADAM- al. 2004) 17 HB-EGF, AR EGFR (Gschwind et al. 2003; Schafer et al. 2004)

Phenylephrine α1AR ADAM- HB-EGF EGFR (Asakura et 12 al. 2002; Yan et al. 2002)

Bombesin BomR ADAM- HB-EGF EGFR (Yan et al. 10 2002)

Carbachol ADAM- Amphiregulin EGFR (Gschwind 17 et al. 2003)

ADAM, a disintegrin and metalloprotease; AT1, angiotensin type 1 receptor; HB, heparin binding; EGFR, epidermal growth factor (EGF) receptors; TGF-α, transforming growth factor-α; GPCR, G protein-coupled receptor; LPA, lysophosphatidic acid; AR, amphiregulin.

Figure 1.18 Schematic representation depicting the proposed TMPS. Multiple GPCRs may induce the activation of MMPs/ADAMs to cleave membrane-bound precursor proteins to release EGFR-ligands which subsequently bind to their respective EGFRs. EGFRs can be activated as homo- or hetero-dimers to initiate signalling pathways that culminates in the stimulation of protein synthesis and cellular hypertrophy. Figure taken from Chan et al.

(Chan et al. 2006).

With regards to the mechanisms by which GPCRs activate ADAM-dependent transactivation, a number of pathways have been identified. Following administration of

α2A-adrenergic GPCR agonist (Pierce et al. 2001) as an example, the dissociation of the trimeric G-protein into βγ-subunits and α-GTP has been proposed to signal to c-src, with subsequent activation of EGFR via HB-EGF shedding (Pierce et al. 2001). The role of src in transactivation appears to be EGFR-ligand dependent. PP1, a well-known src inhibitor has been shown to inhibit BTC-shedding and EGFR transactivation upon glucagon-like peptide 1stimulation of pancreatic cells (Buteau et al. 2003). Another src inhibitor, PP2, does not affect TGF-α shedding, but partially inhibits EGFR transactivation in colonic epithelial cells exposed to carbachol (McCole et al. 2002).

In addition to src, G proteins and PKC, ROS is other intermediate candidate potentially involved in GPCR-transactivation by ADAM stimulation (Gooz 2010). PMA has been shown to activate ADAM-17 through ROS generation in monocytes (Zhang et al. 2001).

Interestingly, in ATII-stimulated COS7 cells, Ca2+ can signal upstream ROS, which in turn activates ADAM-17 to mediate HB-EGF shedding (Mifune et al. 2005). It has been proposed that ROS activates ADAMs/MMPs by oxidising the thiol group from the cysteine residue in the pro-domain, thereby interfering with its association with zinc in the catalytic domain, activating the latent enzyme (Zhang et al. 2001). In the case of ADAM-17 where the inhibitory effect of the cysteine motif in the pro-domain is still dubious, investigations are still underway to understand how latency is maintained.

1.5.5.4 EGFR cross talk with PDGFR

Transactivation of growth factor receptors such as EGFR and PDGFR by stimuli that do not directly bind to the receptor is currently an interesting subject in signal transduction.

Recently, an attractive mechanism for the transactivation of the EGFR has been hypothesised, which suggests metalloproteinase-dependent production of a mature ligand from its precursor (Saito et al. 2001). In addition to its cognate ligands and GPCR, the

EGFR is activated by other stimuli such as RTK, cytokines, chemokines and cell adhesion elements. Moreover, EGFR can be stimulated by UV and gamma radiation, osmotic shock, membrane depolarisation, heavy metal ions as well as radical-generating agents such as

H2O2 (Gschwind et al. 2001).

EGFR transactivation by cytokine receptors has been reported. Like GPCR, cytokine receptors do not possess enzymatic activity, but they can initiate phosphotyrosine- dependent signalling, culminating in MAPK activation (Schafer et al. 2004). Insulin-like growth factor receptor (IGF1R)-activation of EGFR has been reported in COS-7 cells, where IGF1R/EGFR crosstalk relies on MMP-dependent pro-HB-EGF shedding

(Roudabush et al. 2000). In FS4 human skin fibroblasts, PDGF is capable of inducing

EGFR phosphorylation at several distinct sites but could not stimulate phosphorylation of

Thr654, a residue known to be phosphorylated when protein kinase C is activated (Decker and Harris 1989). Likewise, EGFR expression is critical for PDGF-induced migration of murine B82L fibroblasts(Li et al. 2000) . Habib et al. (Habib et al. 1986) were one of the first groups to demonstrate increased PDGFRβ and EGFR association upon PDGF and

EGF stimulation in COS-7, Hs27 fibroblasts, and in a cell line overexpressing the EGFR, 69

R1hER. Alternatively, PDGF can reduce the binding of EGF to EGFR in cells expressing

PDGFR (Li et al. 2000). The decrease in binding seems to occur by a change in EGFR affinity, however the recovery of the normal EGF-binding to EGFR appears to occur 4 h after PDGF exposure (Liu and Anderson 1999). EGFR and PDGFR are concentrated in caveolae of quiescent normal fibroblasts (Liu et al. 1996) hence it is speculated that EGFR transactivation by PDGF occurs within the caveolae. The close proximity of the two receptors is believed to facilitate molecular interactions between the differentially induced signalling pathways (Liu and Anderson 1999).

1.6 Transcription factors

In eukaryotic cells, the regulation of RNA synthesis and subsequent protein expression in response to environmental stimuli is a tightly controlled and regulated process. Induction of different signalling pathways converge to activate or repress target genes. Transcription factors are DNA-binding proteins that recognise cis-regulatory elements in the proximal and distal regions in the promoters of protein-coding genes.

1.6.1 Early growth response factor (Egr-1)

Egr-1, a well-known transcription factor and inducible protein, is a prototype of the immediate early growth response gene family, including Egr-2 and Egr-3. These transcription factors have been implicated in regulating proliferation and differentiation in vascular and other cells (McCafrrey et al. 2000). Importantly, Egr-1 activation does not require de novo protein synthesis, which suggests a role as an immediate response mediator 70

between cell surface receptor signalling and regulation of gene expression regulation

(Khachigian and Collins 1998). Thus, Egr-1 is rapidly and transiently activated by a large number of growth factors e.g. PDGF-BB, cytokines e.g. IL-1β, hypoxia, oxidised lipoproteins, shear stress, ATII, and injurious stimuli (Blaschke et al. 2004).

Egr-1 is an 80 to 82 kDa nuclear phosphoprotein consisting of 533 amino acids. The human Egr-1 gene is located on chromosome 5q23-q31 spanning approximately 3.6 kb

(Silverman and Collins 1999; Blaschke et al. 2004; Khachigian 2006). The DNA-binding domain of Egr-1 consists of 3 zinc fingers of the Cys2-His2 subtype, which are located between amino acids 332 to 416 towards the carboxy-terminal region of the protein (Figure

1.19). These zinc fingers allow Egr-1 to preferentially bind to GC-rich DNA sequences, containing the consensus sequence GCG(T/G)GGGCG (Silverman and Collins 1999). Egr-

1 binding elements have been identified in the promoter region of several genes, including

PDGF, FGF-2, TGF, adhesion molecule ICAM-1, apolipoprotein (A-I), IL-2, p53, CD44,

5-lipoxygenase (LO), cell cycle regulators such as cyclin D1 and retinoblastoma susceptibility gene (Rb), and thymidine kinase, an enzyme integral to DNA biosynthesis

(McCafrrey et al. 2000). Almost all of these genes contain one or more Egr-1 consensus binding sites within their promoter regions (Silverman and Collins 1999).

Figure 1.19 Control elements of the human Egr-1 promoter. The 5’-flanking region contains a TATA box, five serum-response elements (Yang et al.) bearing the sequence

CC(A/T)6GG and binding sites for ternary complex factors (Ets) which belong to the Ets family of transcription factors. The Ets binding sequence is characterised by a conserved

GGAA core. A cyclic AMP response element (Huez et al.) is present. The transcriptional start site is indicated by the arrow. Adapted from Thiel et al. (Thiel et al. 2010).

Transcription of the Egr-1 gene is dependent on RAS-RAF-MEK-ERK1/2 pathway signalling and serum response elements (Yang et al.) in the Egr-1 promoter (Wang et al.

2010). PDGF-induction of Egr-1 mRNA in messangial cells, is associated with phosphorylation of PDGFRβ tyrosines residues and the intact SRE sequences in the Egr-1 promoter (Rupprecht 1993). Furthermore, PDGF-BB-induced Egr-1 activation in SMCs is suppressed by ERK1/2 and src inhibition (Majesky et al. 1992).

Egr-1 shares similar consensus binding sites with other transcription factors such as Sp1 5’-

GGGCGG-3’ and Wilms Tumor suppressor 5’-GNGNGGGNG-3’(Khachigian et al. 1996).

Many promoters contain overlapping Egr-1, Sp1 and WT1 consensus binding motifs.

Studies using recombinant proteins suggest that these transcription factors can displace one another from the binding site of many promoters. An example of this includes regulation of

PDGF-A, in which Egr-1 displaces the pre-bound Sp1 from its binding motif in the promoter (Khachigian and Collins 1997). Egr-1 and Sp1 transcription factors have been shown to activate its transcription (Zhang et al. 2005). Likewise, Sp1 has also been linked to the transcriptional activation of PDGF-B (Khachigian et al. 1994).

In some human tumours, such as breast, brain and lung, fibrosarcoma and gliobastoma,

Egr-1 has been described as a tumour suppressor gene because in these tissues Egr-1 is poorly expressed and can act as growth and transformation suppressor when overexpressed.

Egr-1 can also orchestrate many biological processes from wound repair, female reproductive capacity, controlling synaptic plasticity, inflammation, coagulation, growth 73

control, apoptosis and hyperpermeability of the lung (Thiel et al. 2010). In addition, Egr-1 mRNA and protein levels are up-regulated in glioma cells by EGF (Kaufman and Thiel

2001), in all tissues except the kidney in mice administered EGF (Liu et al. 2000), and in

ECV304 and primary endothelial cells (Tsai et al. 2000). Moreover, Egr-1 pro-tumorigenic transcription factor has been linked to prostate cancer where it is constitutively expressed

(Blaschke et al. 2004). Elevated Egr-1 expression has been correlated with high and impaired activity of p53 in prostate cancer cells (Sauer et al. 2010). Egr-1 in turn, activates the EGFR/ERK signalling cascade, which is known to promote prostate cancer progression

(Sauer et al. 2010). Additionally, Egr-1 silencing in prostate cancer cells has been shown to decrease cell proliferation in vitro (Virolle et al. 2003). Likewise, injection of Egr-1 antisense has been shown to delay tumour growth in TRAMP mice (Baron et al. 2003). It has also been demonstrated that reduced Egr-1 activation correlates with down-regulation of EGFR expression, which in turn results in inhibition of human colon cancer growth

(Chen et al. 2006).

Egr-1 is capable of leading a series of cellular and transcriptional changes and activation/suppression of multiple genes that mediate the development of a variety of pathological processes (Figure 1.20) (Khachigian and Collins 1998). In atherosclerosis,

Egr-1 is weakly expressed in the normal artery wall (Khachigian 2006). In vitro, Egr-1 is strongly up-regulated within minutes in response to a variety of stimuli in SMCs, fibroblasts, leucocytes and ECs (Khachigian and Collins 1998). Following vascular cell activation, Egr-1 is expressed primarily in the nucleus of cells and is capable of activating 74

and controlling the transcription of several genes implicated in the pathogenesis of atherosclerosis and restenosis (Khachigian 2006). McCaffrey et al. (McCafrrey et al. 2000) demonstrated that in human atherosclerotic lesions Egr-1 mRNA levels were elevated compared to the adjacent tissue and that expression was also greater than the constitutive genes such as actin.

Egr-1 is an important mediator of SMC growth and intimal thickening in the reparative response to vascular injury and knock down of Egr-1 proves its role in vascular disease.

For instance, ED5 is a DNAzyme, a catalytic single-stranded DNA that cleaves rat Egr-1 mRNA; targeting the translational start site AUG, at positions 816-817. ED5 suppresses the induction of Egr-1 mRNA and protein and as a result, endothelial cell division, migration and tubule formation in vitro is perturbed (Santiago et al. 1999; Bhindi et al.

2006). ED5 is also able to effectively reduce intimal thickening after balloon injury

(Santiago et al. 1999), or after permanent ligation in rat carotid arteries (Lowe et al. 2002) .

Similar strategies targeting Egr-1 in pig coronary arteries were able to reduce in-stent restenosis (Lowe et al. 2001). Moreover, decoy oligonucleotides targeting Egr-1 inhibit intimal hyperplasia after balloon injury in rabbits (Ohtani et al. 2004).

Figure 1.20 Schematic representation depicting the importance of Egr-1 in various cardiovascular pathological processes. Egr-1 is activated by different pathophysiological stimuli via activation of MAP kinases. Once Egr-1 is induced, it orchestrates the expression of diverse genes implicated in a myriad of cardiovascular conditions such as angiogenesis, ischemia, hypertrophy and lesion development. Other transcriptions factors may also play a role in these processes. Adapted from Khachigian (Khachigian 2006).

Egr-1 DNAzymes have been shown to have anti-angiogenic activity, inhibiting solid tumor growth without compromising wound healing, haemostasis or reproduction (Bhindi et al.

2006). Importantly, Egr-1 DNAzymes inhibit FGF-2 expression, a pro-angiogenic factor downstream of Egr-1, but not VEGF (Fahmy et al. 2003). Consistent with this, ED5 inhibits SMC proliferation by up to 70% (Santiago et al. 1999). Moreover, antisense Egr-1

RNA suppresses the Egr-1 mRNA and protein levels, consequently, SMC proliferation and regrowth in response to injury is significantly inhibited (Fahmy and Khachigian 2002).

Previous studies using Egr-1 DNAzymes demonstrate a reduced myocardial infarct size after ischemia reperfusion in rats (Bhindi et al. 2006), and a decreased neointima formation in pig coronary arteries following stenting (Lowe et al. 2001). The different approaches of inhibiting Egr-1 expression validate the notion that Egr-1 is a master regulator since it controls the expression of multiple growth-regulatory genes. As such, Egr-1 may be an attractive target for gene “knock-down” in a number of pathologies involving excessive cell growth.

Regulatory factors such as p300 and CREB-binding protein (CBP) interact with Egr-1, increasing its transcriptional activation. Other Egr-1 co-activators are NFAT, SF-1, AP-2,

RelA (p65), p53 and Sp1. NAB1 and NAB2 are co repressors of Egr-1 via direct protein- protein interaction. Interestingly, NAB1 is constitutively expressed, while NAB2 is stimulated by known inducers of Egr-1 such as serum and growth factors (Blaschke et al.

2004). It is speculated that NAB interferes with the ability of Egr-1 to bind co-activators including p300 and CBP, thus resulting in its repression or lack of activation (Silverman and Collins 1999). 77

1.6.2 Specificity protein 1 (Sp1)

Sp1 is a member of the Sp family of transcription factors which also includes Sp2, Sp3,

Sp4, BTEB1, TIEG1, TIEG2 and the Kruppel-like factors. Sp1 is divided into two domains, an N-terminal transcriptional activation domain which comprises four different subdomains, A, B, C, and D and a carboxy-termal C2-H2 which contains three contiguous zinc finger domains (Courey and Tijian 1988). Surprisingly, the four Sp1 domains appear to act independently from each other with the exception of domain D (Courey and Tijian

1988) (Figure 1.21).

Sp1 is a 95-105 kDa ubiquitous zinc finger transcription factor; it binds to G-C rich elements (G/A)(G/A)GGCG(G/T)(G/A)G/A) in the promoter of genes (Tan et al. 2008), but it can also bind CAAAT elements (Tan et al. 2008). Sp1 undergoes post-translational modifications; phosphorylation for instance, has been reported to alter Sp1 binding to

DNA. Variations in the DNA-binding capacity of Sp1 may be related to the phosphorylated region on Sp1. For instance, when phosphorylation occurs on the N- terminal region, alterations of gene expression might be expected, without affecting DNA- binding (Chu and Ferro 2005). On the other hand, when Sp1 is phosphorylated in regions responsible for interacting with other proteins, the DNA-binding ability of Sp1 might be altered (Chu and Ferro 2005). Moreover, it has been shown that ERK1/2 phosphorylates

Sp1 on threonine 453 and 739, both in vivo and in vitro, which is associated with up- regulation of VEGF in ECs (Milanini-Mongiat et al. 2002).

Previous studies have shown that Sp1 binds to PDGF-D under basal conditions and ATII exposure (Liu et al. 2006). Likewise, Tan et al. (Tan et al. 2008) demonstrated that residues Thr668, Ser670 and Thr681 in Sp1 are required for Sp1-dependent PDGF-D- activation in response to ATII. In addition, Sp1 has been linked to the transcriptional regulation of PDGF-A (Khachigian et al. 1995) and PDGF-B (Khachigian et al. 1994) in

SMCs.

Figure 1.21 Schematic representation of structural characteristics of Sp1. Coloured boxes indicate regions rich in glutamine (green) and serine/threonine (yellow). +/- refers to regions rich in charged amino acids and AD represents the activation domains. Smaller black boxes represent the three zinc fingers. A, B, C and D domains are also indicated

Figure modified from Suske. (Suske 1999).

1.7 Aims and hypotheses of this thesis

Vascular SMCs are a crucial cell component involved in atherogenesis. Understanding molecular mechanisms controlling this pathological process provides insight into genes orchestrating the disease and consequently identifying potential therapeutic targets. The thesis primarily aimed at elucidating the role of Egr-1, a transcription factor that activates and controls the transcription of several genes implicated in the pathogenesis of atherosclerosis and restenosis. Specific focus was to characterise Egr-1’s role in relation to

PDGF-C expression. This pathway has not been studied in the context of SMC response to

ATII. Furthermore, considering that Egr-1 is a master regulator controlling the expression of multiple growth-regulatory genes, its own transcriptional activation has not been thoroughly investigated.

This thesis was specifically designed to examine the following hypotheses:

A. PDGF-C is upregulated by ATII in SMCs.

1. ATII transiently induces PDGF-C mRNA expression.

2. ATII promotes binding of Egr-1 to the PDGF-C promoter.

3. Egr-1 and Sp1 cooperate in the transactivation of the PDGF-C promoter by

ATII.

4. Mutations of the putative binding site for Egr-1 would abolish promoter

activation of PDGF-C by ATII.

5. Egr-1 is a vital transcription factor in the induction of PDGF-C by ATII.

6. ATII-inducible PDGF-C expression occurs via the AT1R. 81

7. ATII differentially controls the expression of PDGF-C in phenotypically distinct

SMC subtypes.

B. Induction of Egr-1 by PDGF-BB is mediated by novel signalling events in vascular

SMCs.

1. PDGF-BB-inducible Egr-1expression is MMP-dependent.

2. PDGF-BB-inducible Egr-1 expression is ADAM-dependent.

3. MMP/ADAM activity is necessary for PDGF-BB-induced Egr-1 expression in

vitro.

4. PDGF-BB-inducible Egr-1 occurs via the EGFR and the PDGFRβ.

5. Induction of Egr-1 by EGFR-ligands (HB-EGF or EGF) might signal through

the PDGFRβ.

6. PDGF-BB-induced Egr-1 expression does not require other EGFR members

(ErbB2).

7. Egr-1 induction by PDGF-BB does not involve neither ATII receptors nor

FGFR.

8. PDGF-BB promotes physical association between EGFR and PDGFRβ.

9. PDGF-BB-inducible EGFR/PDGFRβ complex is MMP/ADAM dependent.

10. PDGF-BB-inducible SMC migration might be MMP/ADAM/EGFR-dependent.

2 Materials and methods

2.1 Media, buffers and solutions

Waymouth MB752/1 medium (WM)

One vial of WM (WM) MB752/1 medium (Life Technologies, Inc, USA) was mixed with 1

L of dH2O, pH was adjusted to 7. The medium was sterilised by filtration and supplemented with 10 U/mL penicillin, 10 μg/mL streptomycin and 10 % (v/v) foetal calf serum (FCS) from JRH Biosciences (Australia).

TBE buffer (1 x)

21.6 g Tris base (ICN, Biomedical, Inc, France ), 11 g Boric Acid (Sigma, USA), and 80 mL of 0.5 M EDTA (pH 8.0) were made up to 1 L in reverse osmosis water.

RIPA buffer

500 μL of 1x RIPA buffer was prepared by mixing 150 mM NaCl (Sigma, USA), 50 mM

Tris-HCL pH 7.5 (Sigma, USA), 1 % deoxycholate (w/v) (ICN, Biomedical, Inc France),

0.1 % (v/v) SDS, 1 % (w/v) Triton X-100 (Sigma, USA), 1 % (v/v) aprotinin-Trasylol

(Bayer, Germany), 5 mg/mL leupeptin (Sigma, USA), together with 100 mM phenylmethylsulfanyl (PMSF) (Sigma, USA).

4x SDS protein loading sample buffer

500 μL 1 M Tris-HCl (pH 6.8) (Sigma, USA), 800 μL 100 % (v/v) glycerol (Merck,

Germany), 900 μL 20 % (w/v) SDS (INC, Biomedicals, Inc, France), and 400 μL 0.0 5%

(w/v) bromophenol blue (Sigma, USA), were added to 500 μL dH2O.

Resolving gel

x 10 % SDS-PAGE resolving gel was prepared with 2.5 mL of 1.5 M Tris (pH 8.8),

2.5 mL of 40 % (v/v) acrylamide (Bio-Rad Laboratories, Inc), 100 μL of 10 % (w/v)

SDS (INC, Biomedicals, Inc, France), 5 μL TEMED (Sigma), 50 μL of 10 % (w/v)

ammonium persulphate (APS) (Sigma, USA), and 4.86 mL of dH2O.

x 6 % SDS-PAGE resolving gel was prepared with 2.5 mL of 1.5 M Tris (pH 8.8), 1.5

mL of 40 % (v/v) acrylamide (Bio-Rad Laboratories, Inc), 100 μL of 10 % (w/v)

SDS (INC, Biomedicals, Inc, France), 5 μL TEMED (Sigma, USA), 50 μL of 10 %

(w/v) APS (Sigma, USA) and 5.86 mL of dH2O.

Stacking gel (5 % acrylamide)

The stacking gel was made with 1.26 mL of 0.5 M Tris (pH 6.8), 700 μL of 40 % (v/v) acrylamide (BioRad Laboratories, Inc), 50 μL of 10 % (w/v) SDS, 5 μL TEMED, 25 μL of

10 % (w/v) APS and 3.6 mL of dH2O.

SDS running buffer

6 g Tris-base (INC, Biomedicals, Inc, France), 28.8 g glycine (Sigma, USA), 2 g SDS

(INC, Biomedicals, Inc, France), were made up to 1 L with dH2O.

Transfer buffer

6 g Tris-base (INC, Biomedicals, Inc, France), 28.8 g glycine (Sigma, USA), were mixed up with 200 mL methanol and made up to 1 L with dH2O.

PBS (10 x)

One vial of Dulbecco’s Phosphate Buffered Saline (GIBCO ™, Invitrogen Corporation,

USA) was made up to 5 L with dH2O.

PBS-T

0.05 % (v/v) Tween-20 (Sigma, USA) were added to 1 x PBS.

Immunoprecipitation (IP) buffer

IP buffer consisted of 300 μL 5 M NaCl, 100 μL 0.5 M EDTA (pH 8.0), 100 μL 100 %

(v/v) Triton X-100, 5 μL100 % (v/v) NP-40 (Sigma, USA), 500 μL 1M Tris-HCl (pH 7.5),

50 μL 0.1 M DTT and made up to 10 mL with dH2O. Protease inhibitors were also added to the IP buffer: 10 μg/mL leupeptin (Sigma, USA), 0.5 mM PMSF (Sigma, USA), 30 mM p-nitrophenylphosphate (Sigma, USA), 10 mM NaF, 0.1 mM sodium orthovanadate 85

(Na3VO4), 0.1 mM sodium molybdate dehydrate (Na3MoO4) and 10 mM beta- glycerophosphate.

LB-Ampicillin medium

LB growth medium was prepared using 7 LB pellets (BIO101, USA), dissolved in 250 mL of dH2O. Medium was sterilised by autoclaving. For preparation of agar plates, 5 g of bacteriological agar (Gibco) were added prior to autoclaving. For plasmid selection,

Ampicillin (Austrapen, Australia) at concentration of 200 μg/mL, was added to growth medium and plates prior to culture or pouring of plates.

Frozen stocks

0.6 mL of bacterial culture was mixed with 0.4 mL of 50 % (v/v) glycerol (Sigma, USA) and stored at -80 ºC.

Maxi-prep solutions

x Solution I: 9 g of Glucose, 20 mL of 0.5 M EDTA (pH 8.0), 25 mL of 1 M Tris-

HCl (pH 8.0). Made up to 1 L with dH2O.

x Solution II: 40 mL of 1 M NaOH (Sigma), 20 mL of 10 % (w/v) SDS and 140 mL

of dH2O. 86

x Solution III: 45 mL of 3 M potassium acetate (Sigma) (pH 6.8).

TE buffer

10 mM Tris-HCl (pH 7.4), 1 mM EDTA.

Buffers used in nuclear extracts preparation

x Buffer A: 10 mM HEPES (pH 8.0) (Sigma), 1.5 mM MgCl2, 10 mM KCl (Sigma),

0.5 mM DTT, 200 mM sucrose (Sigma), 0.5 % (w/v) Nonidet P-40, 1 μg/mL

aprotinin, 0.5 mM PMSF and 1 μg/mL leupeptin.

x Buffer C: 20 mM HEPES, pH 8.0; 1.5 mM MgCl2, 420 mM NaCl, 0.2 mM EDTA,

1 mM DTT, 1 μg/mL aprotinin, 0.5 mM PMSF and 1 μg/mL leupeptin.

x Buffer D: 20 mM HEPES pH 8.0; 100 mM KCl, 0.2 mM EDTA, 20 % (v/v)

glycerol, 1 mM DTT, 1 μg/mL aprotinin, 0.5 mM PMSF and 1 μg/mL leupeptin.

2.2 Cell culture

Wistar Kyoto 12-22 (WKY12-22) rat neonatal aortic SMC (derived from the aortae of 2- week old pups) and rat adult WKY3M-22 (3-month old) were gifts from Dr. Stephen M.

Schwartz (University of Washington, USA). Primary aortic SMCs (RASMCs) and adult human aortic SMCs (HASMCs) were obtained from Cell Applications Inc (Crawford et al.). Vascular cell lines were cultured in Waymouth’s medium (WM) in 75-cm2 flasks 87

(T75, Nunc, Denmark), and incubated at 37 ºC in a humidified atmosphere of 5 % carbon dioxide (CO2), using an Air Jacket CO2 incubator (Model 60A0100a, Thermoline). At 90

% cell confluency, medium was removed and monolayers rinsed twice with pre-warmed 1 x phosphate-buffered saline (PBS) solution, pH 7.4, followed by addition of 3 mL 0.05 %

Trypsin and 0.02 % ethylenediaminetetraacetic acid (EDTA) in Hank’s balanced salt solution (BioWhittaker) and incubated for 3 min at 37 ºC, 5 % CO2. Flasks were tapped firmly and cells were resuspended in 10 mL, 10 % foetal calf serum-Waymouth (FCS-

WM). The cell suspension was seeded into T75 flasks in 10 mL of medium, gently swirled to ensure proper and even cell distribution, followed by incubation at 37 ºC. WKY12-22 neonatal SMCs were split at a ratio of 1:8, WKY3M-22 SMCs at 1:6, HASMCs and

RASMCs at ratios of 1:4.

2.3 Gene expression analysis

SMCs were subjected to the treatments indicated and mRNA expression levels was evaluated. mRNA levels of a housekeeping gene was used as an internal control.

2.3.1 Total RNA isolation

WKY12-22 and WKY3M-22 SMCs were seeded in 10-cm2 petri dishes in 10 % FCS-WM.

After 48 h, cells were washed twice with pre-warmed 1 x PBS and growth medium was changed to serum free medium and incubated for another 24 h. To determine gene induction by different stimuli, cells were exposed to growth factors for various times. To assess the role of inhibitors on gene induction, a diverse range of pharmacologic inhibitors 88

was used prior to growth factor exposure. Growth factor stimulation was terminated by addition of 1 x ice-cold PBS, immediately followed by removal of PBS and addition of 4.5 mL of TRIzol ® reagent (Life Technologies, Inc, USA) to each petri dish. Cells were lifted with a Costar cell lifter (Corning Incorporated, Mexico), followed by pipetting up and down several times to ensure complete cell lysis. Cell suspensions were transferred to 15 mL falcon tubes. 900 μL of chloroform (Crown Scientific, Australia) was added to each

Falcon tube, vortexed for 30 s and let stand at room temperature (RT) for 5 min, followed by centrifugation for 30 min at 4 ºC at speed of 3800 rpm. The upper phase was transferred to a new 15 mL Falcon tube with addition of 2.5 mL of isopropanol (Crown Scientific,

Australia). Tubes were vortexed for 30 s and left at RT for 1 h to precipitate RNA. Tube contents were centrifuged for 30 min at 4 ºC at 3800 rpm. Supernatant was removed and precipitated RNA was washed with 5 mL of ice-cold 75 % (v/v) ethanol (Crown Scientific,

Australia), followed by centrifugation for 30 min at 4 ºC at 3800 rpm. The supernatant was gently removed and the RNA pellet was air-dried for approximately 1 h. RNA was resuspended in 15 μL RNase free water and stored at -80 ºC. RNA concentration was determined by diluting RNA 1:10 in distilled water (dH2O) and reading absorbance at 260 nm on a Nanodrop 1000 spectrophotometer (Thermo Fisher Scientific, USA). RNA purity was determined by calculating the A260/A280 ratio. RNA with an A260/A280 ratio greater than

1.8 was used for cDNA synthesis.

2.3.2 cDNA synthesis from total RNA

Reaction mix was composed of 5 μg of total RNA, 13 μL of nuclease-free water, 1 μL of

0.5 μg/mLOligo (dT)15 (Sigma), 1 μL of 10 mM deoxyribonucleoside triphosphate mix 89

(dNTPs) (Roche) (10 mM each dATP, dGTP, dCTP and dTTP ) and dH2O to a final volume of 20 μL. Reaction tubes were incubated at 65 ºC for 5 min, immediately chilled on ice then briefly centrifuged. To each sample, 7 μL of the master mix was added. Master mix was composed of 4 μL 5 x First Strand Buffer (Invitrogen, USA), 2 μL of 0.1 M dithiothreitol (DTT) (Invitrogen, USA) and 1 μL of RNA-sin (Promega. USA). The contents were briefly centrifuged and incubated at 42 ºC for 2 min. 1 μL of Superscript II

(Invitrogen, USA) was added and mixed by gently pipetting. Samples were incubated at 42

ºC for 50 min. The reaction was inactivated by heating at 70 ºC for 15 min. Samples were either stored at -20 ºC or immediately used as a cDNA template for amplification in semi- quantitative and/or Real Time polymerase chain reaction (PCR).

2.3.3 Semi-quantitative polymerase chain reaction (PCR)

PCR amplification was performed to evaluate the mRNA expression of PDGF-C and Egr-1 using β-actin expression as a housekeeping control gene. PCR was carried out using the

Platinum Taq polymerase kit (Invitrogen, USA). 1 μL of cDNA was added to a PCR reaction comprising 0.6 μL of 50 mM MgCl2, 2 μL of 10 x PCR buffer, 0.2 μL of 10 μM of each reverse and forward primers (Sigma) (Table 3.1), 0.2 μL of 10 mM of dNTPs (Roche) and 0.2 μL of 1 U Platinum Taq DNA polymerase (Invitrogen) and 15.6 μL of RNase-free water. Cycle-based PCR amplification was performed on a GeneAmp PCR System 2400

(Perkin Elmer, USA).

PCR products were electrophoresed on 1 % (w/v) agarose gels (UltraPureTM Agarose,

Invitrogen) in TBE buffer. Ethidium Bromide 0.1 % (v/v) (Promega ™) was added to melted agarose before the gel was set in a Gel Caster (Bio-Rad Laboratories, USA). A comb was inserted to shape the wells and the agar allowed to solidify at RT. The gel was placed in an electrode chamber (Bio-Rad, Laboratories, USA) with 1 x TBE. Samples and

100 bp DNA ladder (Promega, USA) were loaded and the gel run at 90 V for 1 h. Products were visualised on an UV transilluminator using the Gel Doc 2000 photographic system and Quantity One (version 4.1.0, Bio-Rad).

2.3.4 Real-time PCR (qPCR)

Real-time quantitative PCR was performed using a Corbett RotorGene 3000 Sequence

Detection System (Corbett Life Science, Australia). Reaction mixture contained 10 μL of 2 x SYBR® Green PCR Master Mix (Applied Biosystems, USA), 1 μL of cDNA, 0.3 μL of

10 μM of forward and reverse primers (Sigma) (Tables 3.2 and 4.2) and RNase-free water to a final volume of 20 μL. Readings were acquired at the end of each cycle. PCR product size was verified on a 2 % agarose gel in TBE and sequenced at the Ramaciotti Centre,

(University of NSW, Australia). β-actin mRNA expression served as an internal control.

Levels of β-actin (housekeeping gene) were simultaneously measured with genes of interest for each sample in triplicate. Relative gene expression was calculated using the Delta-Delta

Ct method.

2.4 Western blotting analysis

SMCs were plated onto 10-cm2 petri dishes and grown to 70 % confluency in 10 mL of 10

% FCS-WM. FCS-WM was removed by washing the cells with 1 x pre-warmed PBS. Cell growth was rendered quiescent by incubating in serum-free medium for 24 h, followed by the specified treatment. SMCs were washed twice with 1 x ice cold PBS to stop treatment and cells were lysed with 500 μL of RIPA buffer. Cells were scraped with a Costar cell lifter. Cell lysates were twice frozen down at -80 ºC and then thawed on ice to ensure full disruption. Suspension was centrifuged for 10 min at 4 ºC. The BCA Protein Assay Kit

(Pierce, USA) was used to determine protein concentration according to manufacturer’s instructions. Quantification was performed on a Spectra MAX plus spectrophotometer

(Molecular Devices Corporation, USA) at an absorbance of 562 nm using the SOFT Max

PRO computer software (Version 2.2.1, 1998, Molecular Devices Corporation, USA).

4 x SDS (6 μL) protein loading sample buffer along with 2 μL of 0.5 M DTT and protein lysate were denatured for 5 min by heating at 100 ºC and immediately placing on ice. 2 μL of 0.5 M iodoacetamide (IAM) was added to each tube. Samples were loaded onto SDS-

PAGE gel at the specified percentage of acrylamide. Gels were subjected to electrophoresis at 100 V for 2 or 3 h in SDS running buffer. Kaleidoscope protein ladder

(Promega, USA) was loaded alongside samples (Bio-Rad Laboratories, Inc, USA).

Proteins were transferred onto a presoaked (100% (v/v) ethanol and thoroughly rinsed)

Immobilon-P polyvinylidene fluoride nylon transfer membrane (PVDF) (Millipore, USA) 92

in pre-cooled Transfer buffer containing the “gel-membrane sandwich”, which in a pre- established order consisted of sponge, blotting paper, gel membrane, PVDF, blotting paper and sponge. This was set up in a “sandwich holder” and placed in the electrophoresis apparatus together with an ice block and run at 100 V for 2 h.

After transfer and apparatus disassembly, the gel was stained with GelCode Blue (Pierce,

USA) for 1h and distained in water O/N. The membrane was air-dried, washed with 100 %

(v/v) ethanol and rinsed with dH2O followed by immersion in blocking buffer consisting of

5 % (w/v) skim milk powder in PBS-T. After overnight blocking at 4 ºC with constant gentle shaking, membranes were washed 3 times with PBS-T for 15 min each at RT.

Primary antibodies (Tables 3.3 and 4.3) diluted in 1 % BSA or 5 % (w/v) skim milk-PBS-T

(according to specification sheet) were added to the membranes and incubated for the specified time and temperature. Membranes were then washed 3 times with PBS-T for 15 min each at RT prior to incubation with appropriate secondary antibody conjugated to horse radish peroxidase (HRP) (Dako Corporation, USA). Secondary antibodies were diluted in

5 % skim milk (w/v)-PBS-T. Membranes were washed 3 times with PBS-T for 15 min each. Finally, membranes were incubated with Renaissance chemilumescent reagents

(ECL™ Western Blotting Detection Reagent, GE Healthcare, UK) (mixing equal parts of oxidising and enhanced luminol) for 1 min with gentle swirling. Membranes were twice blotted with blotting paper (to remove excess reagent) and placed between transparent films. Membranes were exposed to X-Ray Hyperfilm (Amershan Life Sciences, UK) for various lengths of time before manual development. 93

2.5 Statistical analysis

Experiments in this thesis were carried out in duplicate on at least two separate occasions.

For Chapter 3, data were analysed for statistical significance using Student’s t-test. For

Chapter 4, statistical comparisons were made by one-way ANOVA with Bonferroni’s

Multiple Comparison Test analysis using GraphPad Prism Version 4.0 (San Diego, CA). A value of p<0.05 was considered statistically significant (*).

2.6 Densitometry

Comparison of the intensity of bands on EMSA and Western Blots was carried out by using the Gel Analysis method outlined in the ImageJ program. The analysed data is plotted as

Relative Intensity of different treatments versus non-treated samples.

2.7 Other methods

Methods specific to the work of individual studies are presented in the relevant chapters.

3 Angiotensin II-induction of platelet-derived growth factor-C

expression is mediated by angiotensin II receptor 1-dependent Egr-1

transactivation

3.1 Introduction and aim

Platelet-derived growth factor-C (PDGF-C) (Li et al. 2000), and PDGF-D are the most recently identified members of the PDGF family of growth factors, which includes the well characterised PDGF-A and PDGF-B. PDGFs are important mitogens, stimulants of motility and differentiation for fibroblasts, SMCs and other cells of mesenchymal origin, but also other cell types including capillary endothelial cells and neurons (Utela et al.

2001). Studies over the last two decades have implicated PDGF-A and -B in pathophysiologic processes such as atherosclerosis, restenosis, fibrosis (Ross 1993; Raines

2004). PDGF-C has been found to contribute to fibrotic disease (Pontem et al. 2003;

Campbell et al. 2005), angiogenesis (Cao et al. 2002; Li et al. 2005), tumorigenesis

(Crawford et al. 2009), embryogenesis (Eitner et al. 2002), palate formation (Ding et al.

2004), platelet activation (Fang et al. 2004). PDGF-C and PDGF-D differ from the other

PDGFs in that these are secreted in a latent form. Proteolytic cleavage of the N-terminal

CUB domain liberates the growth factor domain allowing binding to the PDGF receptors

(PDGFRs) (Lei and Kazlauskas 2008). PDGF-C mRNA is expressed in most human adult tissues and appears to contribute to wound healing by stimulating proliferation of fibroblasts, epithelial migration, vascularisation and neutrophil infiltration (Lei and

Kazlauskas 2008). The highest levels of PDGF-C are found in the heart, liver, kidney, 95

ovary and pancreas (Reigstad et al. 2005), but smaller amounts are also found in the placenta, skeletal muscle and prostate (Li et al. 2000). PDGF-C is overexpressed in atherosclerotic human arteries (Karvinen et al. 2009). Single nucleotide polymorphisms

(SNP) in the PDGF-C gene promoter have been associated with cleft palate (Choi et al.

2009). Functional characterisation of PDGF-C transcriptional activation shows that PDGF-

C is induced by FGF-2, under the transcriptional control of Egr-1, which binds and competes with Sp1 for overlapping binding sites on the PDGF-C promoter, via the ERK1/2 signalling pathway (Midgley and Khachigian 2004).

Of the various inducers of Egr-1, angiotensin II (ATII), the effector peptide of the renin- angiotensin system is a pro-atherogenic factor as it capable of stimulating vascular SMCs proliferation (Mehta and Griendling 2007), that affect the expression of PDGF -A (Day et al. 1999), -B (Deguchi et al. 1999) and -D (Liu et al. 2006; Tan et al. 2008). Vascular

SMC responsiveness to ATII is phenotype-dependent (Griendling et al. 1997). In pup rat medial SMCs, ATII induces PDGF-B mRNA expression, however there is no change in B- chain expression in rat adult SMCs (Deguchi et al. 1999; Khachigian 2006). The two phenotypes bear not only unique markers, but also respond to changes in the extracellular environment through different pathways: (1) the “contractile” phenotype, typical of differentiated arteries, is more responsive to agents that induce vasoconstriction or vasodilatation, and (2) a synthetic phenotype, characteristic of developing and pathologic arteries, is more responsive to growth factors, cytokines and other molecules which induce cell proliferation and migration . 96

ATII activates the extracellular signal-related kinase (ERK) 1/2 to stimulate Egr-1, and its downstream gene PDGF-A, via a G+C-rich region (located -76 to -47 bp) in the proximal region of the PDGF-A promoter, bearing Egr-1 binding element (Khachigian et al. 1995;

Day et al. 1999). This element is similar to that in the proximal PDGF-C promoter

(Midgley and Khachigian 2004). Egr-1 mediates inducible PDGF-A and PDGF-C transcription in cells exposed to FGF-2 through this element (Delbridge and Khachigian

1997; Midgley and Khachigian 2004). Given that this element controls inducible PDGF-A expression in cells exposed to PMA (Khachigian et al. 1995), shear stress (Khachigian and

Collins 1997) and importantly ATII (Day et al. 1999), we hypothesised that this element in the PDGF-C promoter regulates PDGF-C by ATII.

The aim of this chapter was to investigate the effect of ATII on the expression of PDGF-C in two SMC subtypes, in rat neonatal WKY12-22 and in rat adult WKY3M-22 vascular

SMCs. A full understanding of the role of ATII in the expression of PDGF-C would provide important knowledge, which potentially might be used to hamper their pathological role in disease processes such as atherosclerosis. This demonstration is also important, as it would place ATII as an agonist of the four PDGF ligands.

3.2 Materials and methods

3.2.1 Chemicals

ATII was purchased from Sigma. ATII was dissolved in water for irrigation (Baxter), aliquoted and stored at -80 ºC. In various systems, the growth-promoting effects of ATII on mammalian SMCs have been noted with concentrations from 10-7 to 10-5 M (Campbell and Robertson 1981; Herbert et al. 1994; Kohno et al. 2000). Our laboratory has extensive experience in working with ATII at concentration of 10-7 M (Day et al. 1999; Tan et al.

2008). We examined the effect of ATII 10-7 and 10-6 M on WKY12-22 SMCs. The AT1R inhibitor, Losartan, was a generous gift from Merck, Sharp and Dohme (Sydney, Australia).

The AT2R inhibitor, PD123319, was purchased from Sigma.

3.2.2 Cell culture

WKY12-22 and WKY3M-22 vascular SMCs were maintained as described in Chapter 2,

Section 2.2.

3.2.3 RNA preparation and reverse transcriptase reaction

SMCs were grown to 60 % confluence and serum arrested for 24 h prior to exposure to

ATII (10-7 M). In time course experiments, SMCs were exposed to ATII for various times as indicated in the text. In inhibitor studies, 1 μM of Losartan or 1 μM of PD123319 were added for 1 h prior to ATII stimulation for another 2 h. Stimulation was terminated by 98

removal of medium and washing with 1 x ice-cold PBS prior to RNA extraction as described (Section 2.3.1). cDNA synthesis was performed as described (Section 2.3.2).

3.2.4 Semi-quantitative and real-time PCR

PCR amplification was performed to evaluate the mRNA expression of PDGF-C and Egr-1.

β-actin expression was determined simultaneously for normalisation as described (Section

2.3.3 and 2.3.4). Primer details and specific amplification conditions are listed in Tables

3.1 and 3.2 respectively.

Table 3.1 Semi-quantitative PCR conditions and primer sequences used for mRNA gene expression analysis.

Rat PDGF-C (Product size: 481bp) Forward: 5’-GGC ATG AGA GAG TTG TCA CTA TAT CTG GTA-3’

Reverse: 5’-GTC CAA GTC TAT CTG CCA TCG ATC T-3’

Denaturing: 98º C for 1 min, followed by 22 cycles of 96 ºC for 30 s, 60 ºC for 30 s, 72 ºC for 30 s

Extension: 72 ºC for 5 min

Rat Egr- 1 (Product size: 230 bp) Forward: 5’-GCC TTT TGC CTG TGA CAT TT-3’

Reverse: 5’-AGC CCG GAG AGG AGT AAG AG-3’

Denaturing: 98 ºC for 1 min followed by 30 cycles of 94 ºC for 30 s, 60 ºC for 30 s, 72 ºC for 30 s

Extension: 72 ºC for 1 min

Rat Beta-actin (Product size: 228 bp) Forward: 5’-AGC CAT GTA CGT AGC CAT CC-3’

Reverse: 5’-CTC TCA GCT GTG GTG GTG AA-3’

Denaturing: 98 ºC for 1 min followed by 22 cycles of: 94 ºC for 30 s, 58 ºC for 30 s, 72 ºC for 30 s

Extension: 72 ºC for 2 min

100

Table 3.2 Real-time PCR conditions and primer sequences used for mRNA gene expression analysis.

Real time PCR cycling conditions for ALL the genes were as followed:

Initial Hold Tº: 50 ºC for 2 min

Hold Tº: 94 ºC for 10 min

Cycling conditions: 94 ºC for 20 s, 60 ºC for 45 s, 72 ºC for 20 s

Rat Egr- 1 (25 x of cycling conditions) Product size: 230 bp

Forward: 5’-GCC TTT TGC CTG TGA CAT TT-3’

Reverse: 5’-AGC CCG GAG AGG AGT AAG AG-3’

Rat Beta-Actin (25 x of cycling conditions) Product size: 228 bp

Forward: 5’-AGC CAT GTA CGT AGC CAT CC-3’

Reverse: 5’-CTC TCA GCT GTG GTG GTG AA-3’

Rat PDGF-C (40 x of cycling conditions) Product size: 230 bp

Forward: 5’-CAG CAA GTG CAG CTC TCC A -3’

Reverse: 5’-GAC AAC TCT CTC ATG CCG GG-3’

101

3.2.5 Transient transfection

3.2.5.1 Plasmids

The CMV-Sp1 over-expression plasmid was a generous gift from Dr. Robert Tjian

(Howard Hughes Medical Institute, University of California, USA). CMV-gutless was generated in-house by excising the Sp1 DNA sequence from CMV-Sp1 and re-ligating the plasmid. pCB6 and the reporter promoter pCB6-Egr-1 were a generous gift of Dr. Kathleen

Sakamoto (University of California, USA). The pPDGF-C-797 and the pPDGF-C-mut-

Egr-1 promoter construct were generated in the laboratory and previously described

(Midgley and Khachigian 2004). The plasmid control pRL-null, was purchased from

Promega (Crawford et al.).

3.2.5.2 Transformation of competent cells

MAX Efficiency DH5α, Escherichia coli cells (Gibco, USA) were transformed with the relevant plasmids: pPDGF-C-797, pRL-null, pCB6+, pCB6-Egr-1, CMV-gutless, CMV-

Sp1. 50 μL of DH5α cells were thawed on ice. 1 or 2 μg of the required plasmid was mixed with the cells by pipetting and placed on ice for 30 min. The suspension was heat- shocked at 37 ºC for 25 s, immediately placed on ice for 2 min, then transferred into a 15 mL Falcon tube containing 450 μL of pre-warmed S.O.C medium (Gibco, USA). The bacterial culture was agitated in a vertical rotator (Bioline orbital shaker incubator model

BL 4600, Edwards Instrument Company, Australia) for 1 h at 37 ºC prior to spreading on 102

LB/Amp-agar plates (Section 2.1) and amplicillin colonies selected after incubation at 37

ºC overnight.

3.2.5.3 Maxi-prep: large scale DNA purification

A single colony was selected from the Ampicillin selection plates and inoculated into 10 mL of LB/Amp broth (Section 2.1) and placed in a 37 ºC incubator with constant shaking for several hours until turbid. This cell suspension was then poured into 1 L of LB/Amp- broth (Section 2.1) and shaken vigorously at 37 ºC in a vertical rotator overnight. A small aliquot was used to make frozen stocks (Section 2.1). The cell suspension was evenly divided and centrifuged at 5000 rpm in a pre-cooled Sorvall RC-5B Plus centrifuge for 10 min at 4 ºC. After centrifugation the supernatant was decanted and cell pellets resuspended in 40 mL of autoclaved Solution I (Section 2.1), incubated 5 min at RT prior to the addition of 80 mL fresh Solution II (Section 2.1). Tubes were gently mixed by inversion and placed on ice for 15 min. Autoclaved Solution III (40 mL) (Section 2.1), was then added, agitated and centrifuged at 5000 rpm for 30 min at 4 ºC. Supernatants were filtered through sterile gauze and collected into a fresh sterile container, and the pellet discarded. Pre-chilled isopropanol (110 mL) was added and solution centrifuged at 5000 rpm in a pre-cooled

Sorvall RC-5B centrifuge for 20 min at 4 ºC. Supernatants were carefully decanted and discarded and the pellets were washed with 40 mL of ice-cold 70 % (v/v) ethanol, followed by centrifugation at 5000 rpm in a pre-cooled Sorvall RC-5B for 10 min at 4 ºC.

103

Supernatants were discarded and the pellet air-dried for 15 min. Dry pellets were resuspended in 5.5 mL of TE buffer (Section 2.1) before the addition of 6.05 g of cesium chloride (Sigma, USA) (approximately 1.1 g/mL of plasmid solution). Once the cesium chloride was completely dissolved, 0.4 mL of ethidium bromide (10 mg/mL) (Amresc,

USA) was added. Solutions were centrifuged at 3000 rpm for 10 min at RT. Supernatants were loaded into ultracentrifuge tubes (Beckman, USA). Evenly balanced tubes were heat sealed before centrifugation at 55000 rpm overnight at 22 ºC in an ultracentrifuge

(Beckman model L8-55M, USA). DNA bands were withdrawn with an 18 gauge needle attached to a 10 mL syringe. The DNA solution was placed into eppendorf tubes, washed several times with TE-saturated butanol until all trace of red ethidium bromide was removed. 10 mL of TE buffer was added to the colourless layer. 1 mL of 3 M sodium acetate (pH 5.2) (Sigma, USA), and 26 mL of ice-cold 100 % (v/v) ethanol were added.

DNA was precipitated at -20 ºC for a minimum of 30 min and centrifuged at 3000 rpm for

30 min at 4 ºC. Pellets were washed with 70 % (v/v) ice-cold ethanol and re-centrifuged for 15 min at 4 ºC. Supernatants were discarded and the pellets air-dried and resuspended in 1 mL of dH2O. DNA concentration was determined by spectrophotometry (1 μL of

DNA diluted in 69 μL of dH20) at an absorbance of 260 nm.

3.2.5.4 Promoter-dependent expression experiments

Transient transfection analysis of WKY12-22 SMCs using the Dual Luciferase Reporter

(DRL ™) Assay System (Promega, USA) was performed to evaluate PDGF-C promoter- dependent regulation. Briefly, this method employs the Firefly luciferase gene cloned downstream from the PDGF-C promoter and Renilla, to control for transfection efficiency. 104

Firefly and Renilla are two enzymes which are not post-translationally modified but give a luminescent signal when substrate is added. The Firefly luciferase luminescent reaction is measured upon addition of Luciferase Assay Reagent (LARII) which generates and stabilises the signal. The addition of the second reagent, Stop & Glo ® (Promega) quenches the luminescence from the Firefly and activates Renilla.

SMCs were plated in 10-cm2 petri dishes with 10 % FCS-WM and grown to 60 % confluency followed by serum-starvation for 24 h. Transfections were carried out using

FuGENE6 (Roche Applied Science, Germany) at a ratio of 3 μL per μg of DNA, with 5 μg of the reporter plasmid pPDGF-C-797 and 0.5 μg of pRL-null, the Renilla internal control vector. The FuGENE6/DNA mix was incubated for 30 min at RT in 1 mL of serum free medium. The mixture was added drop by drop in a clock-wise manner to each plate before incubation at 37 ºC.

For ATII-treated cells, SMCs were transfected using FuGENE6 with 5 μg of pPDGF-C-

797, 0.5 μg of pRL-null and 2 μg pCB6-Egr-1 or its backbone pCB6+ for 24 h, then stimulated with ATII (10-6 or 10-7 M) for 24 h at 37 ºC and luciferase activity was quantified. For overexpression studies, 5 μg of pPDGF-C-797 and 0.5μg of pRL-null were transfected together with 0, 1 or 2 μg of pCB6-Egr-1 (made up to 4 μg total DNA with its backbone vector pCB6+) or identical amounts of pCB6-Egr-1 with 2 μg CMV-Sp1 (made up to 4 μg with pCB6+). Each transfection was performed in duplicate and incubated at 37

ºC for 24 h. The plates were washed twice with 5 mL of pre-warmed 1 x PBS. 105

Following complete removal of PBS, 1 mL of 1 x Passive Lysis Buffer (DRL ™ Assay

System Promega, USA) was added and the cells detached with cell scrapers. Cell lysates were transferred to Eppendorf tubes and frozen at -80 ºC. Samples were thawed on ice (to aid cell lysing) and spun at 14000 rpm for 2 min. Supernatants were transferred to new tubes. Luciferase activity was quantitated in 10 μL of sample using the Dual Luciferase

Assay System (Promega, USA) containing the Firefly luciferase assay reagent (LAR) and

Renilla reagent Stop & Glo ® as per manufacturer’s instructions. Quantitation of

Luciferase activity was performed in a GloMax® 96 Luminometer (Veritas, Microplate

Luminometer, Turner Biosystems) using white, opaque 96-well opti-plates (Perkin-

Elmer,USA). A total of 50 μL of both Firefly and Renilla substrates per well were added automatically with the luminometer set to 2 s delay before a 10 s activity reading. Results were analysed and graphed using the normalised reading of Firefly luciferase to Renilla luciferase activity.

3.2.5.5 Overexpression studies

In pCB6-Egr-1 and CMV-Sp1 overexpression studies, rat neonatal WKY12-22 SMCs were transfected with 2 μg of pCB6+, pCB6-Egr-1, CMV-gutless or CMV-Sp1. Protein was harvested and collected for Western blotting experiments 24 h after transfection (Section

3.2.6).

106

3.2.5.6 DNAzyme transfection

ED5, the DNAzyme which suppresses Egr-1 expression (Santiago et al. 1999), was custom designed and synthesised by Tri-Link Biotechnologies (CA, USA). A “double hit” transfection (Santiago et al. 1999) using FuGENE6 was performed to evaluate the role of

ED5 or its scrambled counterpart ED5SCR. WKY12-22 SMCs which had reached 80 % confluency were driven to quiescence by replacement of growth medium with serum free medium and cultured for 6 h at 37 ºC. A primary DNAzyme transfection with ED5 at 0.4

μM was performed 18 h prior to stimulation with ATII followed by a second transfection with ED5 (0.4 μM) at the time of exposure to ATII (10-7 M). This double-hit transfection was used to maximise ED5 suppression of Egr-1 gene expression (first transfection should suppress endogenous Egr-1 expression, whilst second transfection should repress the acute induction of Egr-1 by ATII).

3.2.6 Western blot analysis

SMCs were plated onto 10-cm2 petri dishes and grown to 70 % confluency in 10 mL of 10

% FCS-WM. SMCs were serum-starved for 24 h, then exposed to ATII for 0, 0.5, 1 and 2 h or transfected with overexpression plasmids pCB6-Egr-1 and CMV-Sp1(refer to Section

3.2.5.5) for another 24 h. SMCs were washed with 1 x cold PBS to terminate the specified treatments. Western blot analysis was performed as on a 10% SDS-PAGE resolving gel

(Section 2.1). Incubation with primary antibodies Egr-1, Sp1 and beta-actin was done at

RT for 1 h, followed by incubation with their respective secondary antibody. All the secondary antibodies were diluted 1:1000 in 5 % (w/v) SM for 1 h. Primary antibodies are listed in Table 3.3. 107

Table 3.3 List of primary antibodies used in Western blotting experiments.

(a) Dilutions were performed in 0.05 % PBS-T. SM=Skim milk.

Primary Type Dilution Factor Incubation Company

Antibody Resuspended in(a) Cat Number

Egr-1 Polyclonal 1:1000 1 h, RT Santa Cruz

Anti-rabbit 1 % (w/v) BSA sc-189

Sp-1 Polyclonal 1:1000 1 h, RT Santa Cruz

Anti-rabbit 1 % (w/v) BSA sc-59

Beta-actin Monoclonal 1:30000 1 h, RT Sigma

Anti-mouse 5 % (w/v) SM A5316

108

3.2.7 Chromatin immnunoprecipitation assay (ChIP)

ChiP assays were performed utilising human aortic SMCs (HASMCs) grown in 10 -cm2 petri dishes in 10 % FCS-WM until 80 % confluent. Before stimulation with ATII (10-7 M) for 1 h, cells were rendered growth quiescent in serum-free medium for 24 h. HASMCs were washed with 1 x PBS followed by addition of 37 % (w/v) formaldehyde for 15 min before quenching with glycine to a final concentration of 125 mM. Cells were then washed with 1 x PBS, before the addition of Immunoprecipitation (IP) buffer (Section 2.1). Cells lysates were sonicated on ice for six rounds of 15 x 1 s pulses (output level 7) and centrifuged at 14000 rpm for 10 min. Supernatants were evenly fractionated into 4

Eppendorf tubes. Egr-1 (5 μg) and Sp1 (5 μg) antibodies were added to two of them. The four tubes were incubated for 30 min at 22 ºC, 15 min at 4 ºC, followed by centrifugation for 10 min at 14000 rpm. Pre-washed Protein-A and -G beads (Immunoprecipitation

Starter Pack, GE Healthcare, Sweden) were added and the slurry was rotated for 45 min at

4 ºC. The beads were washed five times with pre-chilled IP buffer (Section 2.1) without protease inhibitors, prior to the addition of 100 μL of 10 % (v/v) Chelex-100 (Bio-Rad

Laboratories, Inc) and boiled for 10 min. Samples were then digested with proteinase K

(100 μg/mL) (Invitrogen, USA) for 30 min at 55 ºC, boiled for 10 min at 100 ºC and centrifuged at 14000 rpm for 2 min. Supernatants were collected and subjected to phenol- chloroform extraction and ethanol precipitation (200 μL phenol: chloroform: isoamyl alcohol (25:24:1 saturated with 10 mM Tris (pH 8.0) and 1 mM EDTA). Samples were centrifugated at 14000 rpm for 1 min. The upper layer was collected and 10 μL of 3 M

NAOAc was added, followed by the addition of 2.5 x volume of ice-cooled 100 % (v/v) ethanol. 109

To aid the precipitation of DNA, samples were incubated at -20 ºC for 24 h before centrifugation at 4 ºC for 30 min. Pellets were washed with 70 % (v/v) ethanol and centrifuged at 4 ºC for 10 min. To ensure complete removal of ethanol, DNA was dried by speedivac (Speedvac ® Plus, Sr110A, Savant, USA). DNA was resuspended in 100 μL dH2O and quantified with a Nanodrop 100 (Thermo Fisher Scientific, USA). 1 μL was used in PCR reactions 0.6 μL of 50 mM MgCl2, 2 μL of 10 x PCR buffer, 0.2 μL of 10 μM of each reverse and forward primers (Sigma), 0.2 μL of 10 mM of dNTPs (Roche) and 0.2

μL of 1 U Platinum Taq DNA polymerase (Invitrogen) and 15.6 μL of RNase-free water.

Cycle-based PCR amplification was performed on a GeneAmp PCR System 2400 (Perkin

Elmer, USA). PCR thermal cycling conditions and primer sequences are listed in Table

3.4.

110

Table 3.4 ChIP PCR conditions and primers sequences for amplification of the PDGF-

C promoter containing the putative Egr-1 binding site.

PDGF-C promoter amplification conditions

First Denaturing Tº: 98 ºC for 1 min Denaturing Tº: 98 ºC for 30 s Annealing: 58 ºC for 1 min Repeated 40 times Elongation: 72 ºC for 30 s Final Elongation: 72 ºC for 2 min Forward: 5’-TAG AGG TGT TCC GTG GAA GG-3’ Reverse: 5’-TTG TCC CCT CCC CTT CTC TA -3

Product Size: 330bp

111

3.2.8 Preparation of nuclear extracts

WKY12-22 SMCs were grown in 10 cm2 petri dishes to 60 % confluence, followed by treatments with either ATII (10-7 M) for 1 h or FGF-2 (25 ng/mL) for 4 h. Cells were washed twice with ice-cold PBS (1 x) then scraped into a total volume of 10 mL of cold 1 x

PBS. Cells were then centrifuged for 10 min at 4 ºC, and the pellet resuspended in 100 μL of ice-cold hypotonic solution (Buffer A) (Section 2.1), and incubated on ice for 5 min.

The suspension was centrifuged at 1400 rpm for 40 s and the pellet of nuclei were lysed with 20 μL of an ice-cold solution of Buffer C (Section 2.1) by gently mixing for 20 min at

4 ºC. This mixture was centrifuged at 1400 rpm for 1 min. 20 μL of the supernatant was mixed with 20 μL of cold Buffer D (Section 2.1) and stored at -80 ºC until use.

3.2.9 Electrophoretic mobility shift assay (EMSA)

Binding reactions for gel shift assays were performed with either nuclear extracts (Section

3.2.8) or recombinant proteins (Egr-1 or Sp1). Binding reactions were performed in a total of 20 μL of 10 mM Tris-HCl, pH 8.0; 50 mM NaCl, 1 mM EDTA, 2 mM DTT, 5 % (v/v) glycerol, 1 μg poly (dI-dC), 1 μg salmon DNA (Sigma, USA), γ-32P-ATP (Perkin-Elmer,

100000 cpm) and 6-10 μg of nuclear extracts. The reaction was incubated for 30 min at

RT. In supershift analysis, either 2 μL of Egr-1, Sp1 or ATF-4 antibodies were added to each reaction and incubated with the extracts for 15 min prior to the addition of the probe

(human PDGF-C promoter sequence) 32P-PDGF-C (-560/-518) (5’-CGG GAG AGC CGC

TGA GCC GCC CCC GCT CGC CAG GCG CGC GCT C-3’) or 32P-mEgr-1-P-PDGF-C (-

560/-518) (5’-CGG GAG AGC CGC TGA GCA TAA AAA TAT CGC CAG GCG CGC GCT

C-3’), (Sigma, Genosys, Australia). In experiments using human recombinant proteins, the 112

binding reactions were performed on ice for 15 min, followed by the addition of BSA (2

μg) together with human recombinant Sp1 (100 ng) (Promega, USA) or human recombinant Egr-1 (100 ng) (Alexis Biochemicals,USA). Bound complexes were separated from “free” probe by loading samples onto a 6 % non-denaturing polyacrylamide gel subjected to electrophoresis at 120 V for 2.5 h. The gel was vacuum dried at 80 ºC for 1 h prior to visualisation by autoradiography overnight at -80ºC.

3.2.10 Densitometry analysis

Densitometry analysis were performed as described in Section 2.6.

3.2.11 Statistical analysis

Data was analysed for statistical significance using Student’s t-test, and considered significant when p<0.05. Error bars represent the mean ±SE.

113

3.3 Results

3.3.1 ATII transiently induces PDGF-C mRNA expression in vascular SMC of

neonatal origin

To examine whether ATII induces the expression of PDGF-C in vascular SMCs, quiescent rat neonatal (pups) WKY12-22 SMCs were exposed to ATII for 0.5, 1, 2 or 3 h and harvested for total RNA. PDGF-C chain mRNA was determined by semi-quantitative PCR analysis. Results in Figure 3.1 demonstrate that in WKY12-22 SMCs that ATII (10-7 M) transiently stimulated PDGF-C mRNA synthesis, peaking at 2 h. Beta-actin showed unbiased cDNA loading.

114

Figure 3.1 ATII induces PDGF-C mRNA expression in neonatal SMCs (WKY12-22).

Serum-starved rat neonatal WKY12-22 SMCs were exposed to ATII (10-7 M) for 0.5, 1, 2 or 3 h. (A) Semi-quantitative PCR analysis for PDGF-C gene expression was performed as discussed (Section 3.2.4). Corresponding, beta-actin transcript levels demonstrate unbiased loading. (B) Densitometric analysis of PDGF-C expression relative to beta-actin. NT= no treatment. The data are representative of three different experiments.

(*) denotes p <0.05. Error bars represent the mean ± SE 115

3.3.2 ATII stimulates PDGF-C promoter activity

To determine whether the positive up-regulation of PDGF-C by ATII in WKY12-22 SMCs occurs at the level of transcription, rat neonatal WKY12-22 were exposed to ATII (10-7 M) for 24 h immediately after transfection with pPDGF-C-797, a Firefly luciferase-based reporter construct, bearing 797 bp of the PDGF-C promoter, and pRL-null, a Renilla luciferase-based construct used to normalise transfection efficiency. PDGF-C promoter activity in SMCs transfected with pPDGF-C-797 increased following exposure to ATII (10-

7 M) (Figure 3.2).

116

Figure 3.2 ATII induces PDGF-C promoter activation in neonatal SMCs.

Serum arrested rat neonatal WKY12-22 SMCs were transfected with 5 μg of the reporter

Firefly reporter construct pPDGF-C-797 together with 0.5 μg of the Renilla control pRL- null prior to treatment with ATII (10-7M) for 24 h. Firefly activity was normalised to

Renilla activity. Data are representative of two separate determinations and are shown as relative arbitrary units A value of p<0.05 was considered as statistically significant (*).

Error bars represent the mean ± SE. (Assay performed by Dr. Valerie Midgley).

117

3.3.3 Egr-1/Sp-1 binding site (-35/-1 bp) in the PDGF-C promoter is not required

under ATII exposure

Studies performed previously by our laboratory demonstrated that ATII induction of

PDGF-A expression is mediated via ERK1/2-dependent Egr-1 binding to a G+C-rich region of the PDGF-A promoter at overlapping Egr-1/Sp1 binding sites (-76 to -47 bp)

(Day et al. 1999). This site is remarkably similar to that found in the proximal region of the PDGF-C promoter. Hence, to determine whether ATII induces Egr-1 binding to this region of the PDGF-C promoter, an electrophoretic mobility shift analysis (EMSA) was performed using nuclear extracts from rat neonatal WKY12-22 SMCs which had been treated with or without ATII (10-7 M) for 1 h and a 32P-labeled double-stranded oligonucleotide probe whose sequence spans from -35 to -1 bp of the PDGF-C promoter

[32P-Oligo C (-35/-1)]. This sequence was previously shown to bind FGF-2-inducible Egr-1

(Midgley and Khachigian 2004).

EMSA provided evidence for the formation of four nucleo-protein complexes, three major

(Na, Nc, Nd) and one minor (Nb) (Figure 3.3A). Specific involvement of a transcription factor in each nucleo-protein complex is indicated by supershifting of bands in the presence of the relevant antibody, while a lack of changes in the binding pattern indicates nonspecificity. Incubation of SMCs nuclear extracts with antibodies to Sp1 at room temperature for 20 min eliminated complex Na indicating that Sp1 binds to this region of the PDGF-C promoter (Figure 3.3A). In contrast, antibodies to Egr-1 and ATF-4, under the same incubation conditions had no effect (Figure 3.3A). 118

The lack of band shift with the Egr-1 antibody was surprising given that previous studies demonstrated that ATII induced Egr-1 expression in SMCs (Ling et al. 1999). It was therefore necessary to ensure that in our cellular model, ATII also induced Egr-1 expression and to establish the integrity of the Egr-1 complex. Semi-quantitative PCR and western blot analyses were performed on mRNA and protein extracts of quiescent rat neonatal

WKY12-22 SMCs exposed to ATII (10-7 M) for 0.5, 1, 2 or 3 h. Semi-quantitative PCR analysis revealed that ATII up-regulated Egr-1 mRNA as early as 30 min after exposure

(Figure 3.3C), which was corroborated at the protein level (Figure 3.3D). These findings demonstrate that although ATII induces Egr-1 mRNA and protein expression in SMCs,

ATII does not influence the binding of Egr-1 to the proximal region of the PDGF-C promoter (-35 to -1). To establish the integrity of the Egr-1 complex, EMSA was performed using nuclear extracts of rat neonatal WKY12-22 SMCs treated with FGF-2 (25 ng/mL) together with 32P-Oligo-C (-35/-1). As previously shown (Midgley and Khachigian

2004), this produced six nucleoprotein complexes (N1-N6) (Figure 3.3B). Super shift analysis demonstrated that complex N3 contained Egr-1 and N1 contained Sp1.

119

A B

C D

120

Figure 3.3 ATII does not induce binding to the Egr-1/Sp1 binding site (-35 to -1) on the PDGF-C promoter despite up-regulating Egr-1 mRNA and protein levels.

(A) EMSA using 32P-Oligo-C spanning the -35/-1 region of the PDGF-C promoter together with ATII-treated (10-7 M) or untreated nuclear extracts from rat neonatal WKY12-22

SMCs. Supershift analysis was performed by incubation with Sp-1, Egr-1 and ATF-4 antibodies. Na, Nb, Nc, Nd: nucleoprotein complexes formed. (B) EMSA using 32P-Oligo-

C spanning the -35/-1 region of the PDGF-C promoter in conjunction with nuclear extracts from rat neonatal WKY12-22 SMCs treated with FGF-2 (25 ng/mL) for 1 h. Supershift analysis was performed by incubation with Sp-1 or Egr-1 antibodies to demonstrate the identity of the nucleoproteins complexes (N1, N2, N3, N4, N5 and N6). (C) Semi- quantitative PCR and (D) Western Blot analyses of neonatal WKY12-22 SMCs treated with ATII (10-7 M) for the indicated times.

121

3.3.4 A novel Egr-1 binding site motif in the PDGF-C promoter (-543/-535) acts as a

functional Egr-1-response element in WKY12-22 SMCs

More detailed examination of the proximal region of the PDGF-C promoter identified another putative Egr-1 binding element (5’-CGCCCCCGC-3’) present at position -543/-

535 relative to the transcriptional start site (Figure 3.4A and B). To determine whether

ATII induces binding to this putative element, an EMSA was performed using nuclear extracts of rat neonatal WKY12-22 SMCs treated with or without ATII (10-7 M) for 1 h and a 32P-labeled-PDGF-C(-560/-518) probe, whose sequence spans the Egr-1 putative element -

543/-535. Two nucleoprotein complexes (Ni and Nii) were produced when untreated-ATII nuclear extracts were incubated with 32P-labelled-PDGF-C (-560/-518) oligonucleotide (Figure

3.5A, Lane 2 and Figure 3.5B). Sp1 antibody supershifted the Ni complex, whereas the

Egr-1 antibody reduced its intensity (Figure 3.5A, Lanes 4 and 5, and Figure 3.5B). As expected, the irrelevant antibody control ATF-4, had no significant effect on the complex

(Figure 3.5A, Lane 6 and Figure 3.5B). Complex Ni appears to involve binding of both

Egr-1 and Sp1, with Sp1 as a predominant occupant. Interestingly, a second complex, Nii, was also induced upon ATII exposure (Figure 3.5A, Lane 3 and Figure 3.5B). The Nii complex was abrogated after incubation with Egr-1 antibody, suggesting involvement of the later in this complex (Figure 3.5A, Lane 4 and Figure 3.5B).

122

123

Figure 3.4 Proximal region of the PDGF-C promoter.

(A) A consensus Egr-1 binding element was found at -543 to -535 bp in the PDGF-C promoter, upstream of the G+C rich element (-35 to -1bp). (B) Sequence of the human

PDGF-C promoter region. Egr-1 consensus elements are underlined in black. The EMSA probe containing this element is denoted by blue italics. The primer sequences used for

ChIP experiments are denoted in green bold type with the respective arrows depicting forward and reverse sequences. The relative base pairs are represented by numbers in bold type and the curved arrow at the bottom (right-hand) represents the putative transcriptional start site as defined by UCSC Genome Database.

124

To determine the nucleotides critical for Egr-1/Sp1 binding to the PDGF-C promoter, a transverse mutation was introduced into 32P-PDGF-C (-560/-518), mutating the putative

Egr binding site -543CGCCCCCGC-535 to -543ATAAAAATA-535. EMSA was performed utilising this mutated oligonucleotide 32P-mEgr-1-PDGF-C (-560/-518) with nuclear extracts of neonatal WKY12-22 SMCs treated with or without ATII (10-7 M) for 1 h. Mutation of the consensus Egr-1 motif in the oligonucleotide abrogated complex formation in both stimulated and unstimulated SMCs (Figure 3.5A, Lanes 8 and 9) demonstrating that the integrity of the -543/-535- region is vital for the interaction between Egr-1 and Sp1 with this region of the PDGF-C promoter.

These results demonstrate that Sp1 interacts basally with the -560/-518 site in the PDGF-C promoter and that Egr-1 inducibly binds to it upon ATII stimulation. Chromatin immunoprecipitation (ChIP) analysis using Egr-1 and Sp-1 antibodies further supported the findings (Figure 3.6), that Sp1 associates with the -560/-518 region in the PDGF-C promoter under basal conditions and that ATII increases Egr-1 occupancy of the PDGF-C promoter at the-543/-535 site.

125

Nii

126

Figure 3.5 ATII induces Egr-1 binding to the PDGF-C promoter.

(A) EMSA was performed using nuclear extracts from neonatal WKY12-2 SMCs treated with or without ATII (10-7 M) for 1 h and a 32P-labeled-PDGF-C (-560/-518) or 32P-mEgr-1-

PDGF-C (-560/-518)probe. EMSA Experiment performed by Dr. Valerie Midgley. (B)

Quantitation of band intensities (complexes Ni and Nii) in the NT/ATII-treated nuclear extracts incubated with 32P-labeled-PDGF-C (-560/-518) by scanning densitometry. Statistical studies were performed by using student’s t-test. (*) represents p<0.05 which was considered as statistically significant.

127

Figure 3.6 ATII induces Egr-1 binding to the PDGF-C promoter by ChIP.

ChIP analysis with untreated and treated human aortic SMCs extracts pulled down with anti-Sp1 or anti-Egr-1 antibodies. A region of the PDGF-C promoter spanning Egr-1 motif

-543/-535 was amplified by PCR. The amplicons were confirmed by nucleotide sequencing. The “No Ab” control indicates no antibody added to the extracts.

128

3.3.5 Egr-1 and Sp-1 interact and bind to the PDGF-C promoter

Sp1 and Egr-1 are known to physically interact (Lin and Leonard 1997). In order to confirm the interaction of Sp1 and Egr-1 within the PDGF-C promoter (-560/-518), an

EMSA experiment was performed with 32P-labeled-PDGF-C (-560/-518) and human recombinant Sp1 and Egr-1. A nucleoprotein complex was not formed in the presence of either recombinant Sp1 or Egr-1 when added individually (Figure 3.7, Lanes 2 and 3).

However, a single complex was produced when both Sp1 and Egr-1 were co-incubated with the 32P-labeled-PDGF-C (-560/-518) oligonucleotide. This supports the previous EMSA and

ChIP results which indicated that the presence of these two transcription factors is necessary for the formation of a nucleoprotein complex. Interestingly, only one major complex formed when 32P-labeled-PDGF-C (-560/-518-) was incubated with both Sp1 and Egr-

1 recombinant proteins (Figure 3.7, Lane 4). This complex is similar to the Ni nucleoprotein complex (containing both transcription factors) observed in the ATII-treated

WKY12-22 SMC nuclear extracts.

129

Figure 3.7 Egr-1 and Sp1 interact and bind to the PDGF-C promoter.

EMSA was performed with 32P-labeled-PDGF-C (-560/-518-) together with human recombinant Sp1 (100 ng) (Lane 2), Egr-1 (100 ng) (Lane 3), or both Egr-1 and Sp1 (100 ng each) (Lane 4). Arrow indicates the complex formed. Assay performed by Dr. Valerie

Midgley.

130

3.3.6 Egr-1 and Sp1 cooperatively transactivate the PDGF-C promoter

To provide functional evidence for the influence of Egr-1 and Sp1 on PDGF-C promoter activity, overexpression studies in conjunction with promoter-dependent reporter experiments were carried out. Transfection of WKY12-22 SMCs with low amounts of pCB6-Egr-1 (1-2 μg) resulted in modest activation of the PDGF-C promoter (Figure 3.8, comparing bars 2 and 3 versus 1). In the presence of a constant concentration of CMV-

Sp1 (2 μg), together with pCB6-Egr-1 (1-2 μg), luciferase activity of the PDGF-C promoter was significantly increased (Figure 3.8, comparing bars 5 and 6 versus 4). These results support the concept of cooperativity between Egr-1 and Sp1 in inducing PDGF-C promoter activity.

131

Figure 3.8 Egr-1 and Sp1 cooperatively induce PDGF-C promoter activity.

Growth arrested rat neonatal SMCs were transfected with 0, 1 or 2 μg of pCB6-Egr1 (made up to 4 μg with its backbone pCB6+ plasmid), identical amounts of pCB6-Egr-1 (0, 1 or 2

μg), 2 μg CMV-Sp1 (made up to 4 μg with the backbone pCB6+ plasmid) or 2 μg of each backbone plasmid (pCB6+ and CMV-gutless), together with 5 μg of the pPDGF-C-797 reporter plasmid and 0.5 μg of pRL-null. Luciferase activity of the PDGF-C promoter was measured 24 h after transfection. Firefly luciferase activity was normalised to Renilla activity to correct for transfection efficiency. Data are represented as fold change relative to the backbone-transfected SMC (pCB6+ or CMV-gutless). The result is representative of at least three independent observations. (*) represents p<0.05 by t-test statistical analysis.

Error bars represent the mean ±SE. 132

3.3.7 Egr-1/Sp1 binding element (-543/-535-) in the PDGF-C promoter is required

for ATII-inducible PDGF-C transcriptional activation

Having shown that Egr-1 and Sp1 do not bind to the mutated form of the -543/-535 response element in the PDGF-C promoter and that mutation of this response element failed to support formation of either the Ni or Nii complexes, the same mutation was introduced into the pPDGF-C-797 luciferase reporter construct. This construct (pPDGF-C-

797-mutEgr-1) fortuitously harboured this region, and the effect of this mutation on ATII- inducible PDGF-C luciferase activity was evaluated. Consistent with the earlier findings, transient transfection analysis revealed that ATII-stimulated wild type PDGF-C luciferase activity in neonatal WKY12-22 SMCs over 7-fold after 24 h treatment (Figure 3.9). In contrast, in WKY12-22 SMCs-transfected with the mutant construct, pPDGF-C-797- mutEgr-1, ATII no longer up-regulated PDGF-C promoter-dependent expression and remained at levels comparable to un-stimulated WKY12-22 SMCs transfected with pPDGF-C-797-mutEgr-1. These results further confirm that the integrity of the 543/-535 response element in the PDGF-C promoter is vital for ATII-driven activation of the PDGF-

C promoter.

133

Figure 3.9 Novel Egr-1 response element in the PDGF-C promoter is responsible for

ATII-inducible PDGF-C expression.

Transient transfection analysis was performed in rat neonatal WKY12-22 SMCs using 5 μg of pPDGF-C-797 or 5 μg of pPDGF-C-mutEgr-1 and 0.5 μg of pRL-null stimulated with or without ATII (10-7 M) for 24 h. Firefly and Renilla luciferase activity was determined in cell lysates. Firefly luciferase activity was normalised to Renilla activity to correct for transfection efficiency. (*) denotes p< 0.05 by Student’s t-test analysis. Assay performed by Dr. Valerie Midgley.

134

3.3.8 Egr-1 DNAzyme blocks ATII-inducible PDGF-C expression

Having demonstrated that Egr-1 inducibly binds to a novel binding element in the PDGF-C promoter upon ATII stimulation, it was hypothesised that Egr-1 is required for ATII induction of PDGF-C. To determine the effect of Egr-1 knockdown on ATII-inducible

PDGF-C mRNA expression, rat neonatal WKY12-22 SMCs were transfected with the

DNAzyme ED5 (Khachigian 2000; Santiago and Khachigian 2001), a catalytic oligodeoxynucleotide that binds to and cleaves the Egr-1 mRNA. Alternatively, rat neonatal WKY12-22 SMCs were also transfected with a scrambled control, ED5Scr, consisting of an active catalytic domain but scrambled RNA-hybridising arms. A “double- hit” transfection was performed (Section 3.2.5.6). This protocol has been shown to result in >75% uptake in our laboratory as determined by fluorescence microscopy with transfected FITC-labelled DNAzyme (Santiago et al. 1999).

PDGF-C mRNA expression was assessed 2 h after ATII exposure of SMC. Semi- quantitative PCR (Figure 3.10A) and real-time-PCR (Figure 3.10B) analyses show that

ED5 down-regulated ATII-inducible PDGF-C mRNA, whereas the ED5Scr had no effect.

These results were consistent with the reduction in ATII-inducible Egr-1 mRNA expression by ED5 but not ED5Scr (Figure 3.11A and B). Together, these findings demonstrate that

PDGF-C up-regulation by ATII is Egr-1-dependent.

135

Figure 3.10 Egr-1 DNAzyme blocks ATII-inducible PDGF-C expression.

Rat neonatal WKY 12-22 SMCs were double transfected over an 18 h period with the

DNAzyme ED5 (0.4 μM) or its scrambled counterpart ED5Scr (0.4 μM) and incubated with or without ATII (10-7 M) for 2 h. Assessment of PDGF-C mRNA levels was carried out by semi-quantitative-PCR (A) Beta-actin was used to show unbiased loading. (B) Real-time

PCR analysis investigating the effect of ED5 on ATII-inducible PDGF-C mRNA expression. (*) denotes p< 0.05 by Student’s t-test analysis. Data are representative of two or more independent determinations.

136

Figure 3.11 Egr-1 DNAzyme blocks ATII-inducible Egr-1 expression.

Rat neonatal WKY 12-22 SMCs were double transfected over an 18 h period with the

DNAzyme ED5 (0.4 μM) or its scrambled counterpart ED5Scr (0.4 μM) and incubated with or without ATII (10-7 M) for 2 h. Assessment of Egr-1 mRNA levels was carried out by semi-quantitative-PCR (A) Beta-actin was used to show unbiased loading. (B) Real-time

PCR analysis investigating the effect of ED5 on ATII-inducible Egr-1 mRNA expression.

(*) denotes p< 0.05 by Student’s t-test analysis. Data are representative of two or more independent determinations. 137

3.3.9 Egr-1 overexpression potentiates the activation of the PDGF-C promoter by

ATII

Building on the previous results showing that ATII-activates PDGF-C promoter activity in

SMCs (Section 3.3.2) and the dependency of ATII-induction of PDGF-C mRNA levels on

Egr-1 (Section 3.3.6 and 3.3.8), it was explored whether exogenous Egr-1 affects PDGF-C promoter-dependent expression upon ATII stimulation. Rat neonatal WKY12-22 SMCs were co-transfected with pCB6-Egr-1 (2 μg), or its backbone control (pCB6+), and a Firefly luciferase-based reporter construct pPDGF-C-797 (5μg) and incubated with or without

ATII at different concentrations (10-6 M or 10-7 M) (Figure 3.12). Luciferase activity 24 h after ATII stimulation revealed that overexpressed Egr-1 potentiated ATII-inducible

PDGF-C promoter activation. Interestingly, this response was in an ATII dose-dependent manner. This result confirms that Egr-1 is critical for ATII-inducible PDGF-C expression in these SMCs and acts at the level of promoter activation.

138

Figure 3.12 Egr-1 positively influences ATII-inducible activation of the PDGF-C promoter.

Growth-quiescent WKY12-22 SMC were co-transfected with the PDGF-C promoter reporter construct pPDGF-C-797 (5 μg), together with pCB6+ or pCB6-Egr-1 (2 μg) and pRL-null (0.5 μg), and incubated overnight. SMCs were exposed to ATII (10-6 M or 10-7

M) for another 24 h. Firefly and Renilla luciferase activity was determined in cell lysates.

Firefly luciferase activity was normalised to Renilla activity to correct for transfection efficiency and measured in a luminometer. The result is representative of at least two independent observations. (*) denotes p<0.05 by Student’s t-test statistical analysis. Error bars represent the mean ±SE.

139

3.3.10 Angiotensin II stimulates PDGF-C mRNA expression via Angiotensin II

receptor 1

ATII, a product of the rennin-angiotensin system, is known to stimulate physiologic systems that increase blood pressure and pathological conditions associated with vascular

SMC growth (Khachigian et al. 2000). ATII acts via two high-affinity receptor subtypes, designated as AT1R and AT2R (Berk 2003). The induction of PDGF-A (Day et al. 1999),

PDGF-B (Deguchi et al. 1999; Ishizaka et al. 2006) and PDGF-D (Ishizaka et al. 2006) by

ATII occurs through AT1R, however whether ATII stimulates PDGF-C expression through a specific receptor in neonatal SMCs is unknown. Serum-starved WKY12-22 SMCs were incubated with Losartan (1 μM), a competitive inhibitor of AT1R (Day et al. 1999), or the

AT2R antagonist, PD123319 (1 μM) (Deguchi et al. 1999), for 1 h prior to stimulation with

ATII (10-7 M) for 2 h. The effect of the pharmacologic inhibitors on ATII-inducible PDGF-

C mRNA levels was assessed by real-time-PCR (Figure 3.13). Losartan blocked ATII- inducible PDGF-C expression, whereas PD123319 had no effect, demonstrating that ATII employs the AT1R to initiate its signalling activation which ultimately leads to PDGF-C gene expression.

140

Figure 3.13 ATII-inducible PDGF-C expression acts through the AT1R.

Real-time PCR analysis of cDNA samples from growth-quiescent rat neonatal WKY12-22

SMCs, pretreated for 1 h with Losartan or PD123319 (1 μM each) before incubation with

ATII (10-7 M) for 2 h. PDGF-C expression has been normalised to beta-actin mRNA levels to correct for cDNA concentration between samples. (*) denotes p<0.05. Error bars represent the mean ±SE. Experiment performed by Dr. Valerie Midgley.

141

3.3.11 Egr-1-induction of PDGF-C transcriptional activation is abrogated upon

mutation of a putative Egr-1 motif (-543/-535) in the PDGF-C promoter

To extend on the functional importance of the novel upstream element (-543/-535) in relation to Egr-1 and Sp-1 transcriptional activation of the PDGF-C promoter, transient co- transfection studies were carried out in WKY12-22 SMCs using wild-type (pPDGF-C-797) and a mutant form (pPDGF-C-797-mut Egr-1), together with Egr-1 and Sp-1 overexpression plasmids and their respective backbones. As expected, both transcription factors activated the wild-type PDGF-C promoter (Figure 3.14A, bars 2 and 4).

Interestingly, whereas the mutated PDGF-C promoter was no longer activated by Egr-1

(Figure 3.14A, bar 6), disruption of the -543/-535 response element did not block the ability of Sp1 to activate the promoter (Figure 3.14A, bar 8). Overexpression of Egr-1 and

Sp1 were confirmed by western blot analysis. Egr-1 and Sp1 protein expression was induced by pCB6-Egr-1 and CMV-Sp1 respectively, compared to their backbones alone

(Figure 3.14B and C). These findings demonstrate that Sp1 transactivation of the PDGF-C promoter does not occur via the -543/-535 site, even though Sp1 binds to this element.

142

B C

Figure 3.14 Mutation of the -543/-535 motif represses Egr-1 activation of the PDGF-C promoter.

(A) Transient co-transfection in WKY12-22 SMCs with 5 μg of pPDGF-C-797 or pPDGF-

C-797-mut Egr-1 together with 2 μg of pCB6-Egr-1 (or its backbone, pCB6+) or 2 μg of

CMV-Sp1 (or its backbone, CMV-gutless). Firefly and Renilla luciferase activity was determined in cell lysates. Firefly luciferase activity was normalised to Renilla activity to correct for transfection efficiency. (*) denotes p<0.05 using Student’s t-test statistical analysis. Error bars represent the mean ± SE. (B) Western blot analysis of Egr-1 and Sp1 in transiently transfected WKY12-22 SMCs with pCB6-Egr-1 or CMV-Sp1 or with respective backbones. 143

3.3.12 ATII differentially regulates PDGF-C in phenotypically distinct SMC subtypes

SMCs are not a terminally differentiated cell type and display large phenotypic diversity and plasticity which allows for a broad range of functions in physiological processes and pathological states (Yoshida and Owens 2005), by allowing them to rapidly adapt to changing environmental cues (McDonald and Owens 2007). In order to explore whether

PDGF-C induction by ATII is evident in other SMC subtypes, rat adult WKY3M-22 SMCs were exposed to ATII for 0.5, 2, 4 or 8 h. PDGF-C mRNA expression assessed by semi- quantitative PCR, shows that ATII at (10-7 M) did not increase PDGF-C transcripts in this

SMCs (Figure 3.15A), in contrast to that found in neonatal SMCs (Figure 3.1), indicating that PDGF-C expression in SMC is phenotype-depedent. Consequently, to explore whether the inability of ATII to influence PDGF-C expression in rat adults WKY3M-22 is due to the lack of induction of Egr-1 by ATII in this cell type, Egr-1 mRNA expression was determined by semi-quantitative PCR. Rat adults WKY3M-22 were exposed to ATII (10-7

M) for 0.5, 2, 4 or 8 h. Semi-quantitative PCR (Figure 3.15B) analysis revealed that ATII stimulates Egr-1 mRNA expression in rat adult WKY12-22 SMCs in a time-dependent manner as was found in rat neonatal WKY12-22 SMCs (Figure 3.3C and D). Overall, the findings indicate Egr-1-dependent and -independent transcriptional control of PDGF-C between these SMC subtypes.

144

Figure 3.15 ATII does not influence PDGF-C mRNA levels in rat adult aortic SMCs despite of inducing Egr-1 mRNA expression.

(A) PDGF-C mRNA and (B) Egr-1 mRNA expression were determined by semi- quantitative PCR analysis after ATII (10-7 M) stimulation of rat adults WKY3M-22 for 0.5,

2, 4 or 8 h. Beta-actin mRNA levels were used as loading control.

145

3.4 Discussion

The present study provides the first evidence that ATII can significantly induce PDGF-C expression in neonatal SMCs (WKY12-22), therefore defining ATII as an agonist of all four known PDGF-ligand chains. Given that ATII is an important hormone implicated in the abnormal growth of SMCs (Kuma et al. 2007), and that PDGF-C is expressed in SMCs of the artery wall (Fang et al. 2004), the present findings are important to further understand the role and the relationship of these molecules in the development of atherosclerosis.

Mechanistically, this study has defined that ATII-induction of PDGF-C in neonatal SMCs occurs at the level of transcription, activating its promoter and up-regulating mRNA expression levels. Further, this study has demonstrated that ATII-induction of PDGF-C expression in neonatal SMCs is under the regulatory influence of Egr-1, via a new Egr-1 binding site in the PDGF-C promoter (-543/-535), located upstream of the transcriptional start site. This novel upstream element was shown to be responsible for cooperatively binding both Egr-1 and Sp1 in the presence of ATII. Previous studies have demonstrated that FGF-2 can up-regulate PDGF-C in neonatal SMCs via an Egr-1/Sp1 element (-35 to -1 bp) (Midgley and Khachigian 2004). DNAzymes targeting Egr-1 mRNA (ED5) established that Egr-1 plays a pivotal role in the ATII-inducible PDGF-C expression in rat neonatal

WKY12-22 SMCs. Previously, ED5 but not its scrambled counterpart was shown to block

FGF-2-inducible PDGF-C mRNA expression in neonatal SMCs (Midgley and Khachigian

2004). This study also confirms the capacity of ATII to regulate the expression of the zinc 146

finger transcription factor Egr-1, previously shown to occur in rat vascular smooth muscle cells (Sachinidis et al. 1992).

Additionally, by using specific inhibitors of ATII receptors, AT1R and AT2R, it was demonstrated that ATII acts through AT1R to induce Egr-1 and PDGF-C expression.

Properties or gene profile of pups WKY12-22 SMCs has been shown (Lemire et al. 1994).

Unfortunately, no information on expression of ATII receptors was mentioned. The AT2R is highly expressed in foetal tissues and its expression decreases after birth. The AT2R is re-expressed in the adult animal after vascular injury and wound healing (Nouet and

Nahmias 2000). It is well known that the AT2R blocks ERK activity in vitro and in vivo and promotes apoptosis (Huang 1996; Horiuchi 1997; Akishita 1999). The result obtained in this thesis, demonstrating that ATII-inducible PDGF-C expression occurs via AT1R needs to be carefully interpreted. Due to the synthetic phenotype of WKY12-22 SMCs and their greater responsiveness to injury stimuli (Chapter 3), the stimulation of these cells with

ATII might have resulted in down-regulation of the AT2R but up-regulation of the AT1R.

In this is shown by the lack of effect of the AT2R inhibitor P123319 on ATII-induced

PDGF-C expression in contrast to the blockade of AT1R by Losartan. Importantly, gene transfer studies have shown that AT2R expression antagonizes the proliferative and migratory roles of AT1R in SMCs exposed to growth factors (Nakajima et al. 1995). On the other hand, based on the synthetic phenotype (highly migratory and proliferative) of pups SMCs, it could be hypothesized that this cell line does not express AT2R. In vivo studies in mice with disruption of AT2R have demonstrated exacerbated neontimal formation compared with wild type mice (Suzuki et al. 2002). Consequently, further 147

studies on expression pattern of ATII receptors in this cell line would assist in the understanding of the role these receptors play in ATII signalling.

The present study also demonstrates that ATII-inducible PDGF-C expression in SMCs is differentially regulated in cells of neonatal and adult origin. Phenotypically diverse SMCs are found in vivo. Of utmost interest, is the knowledge of the differences between intimal and medial SMCs in the adult animal and importantly differences observed between subpopulation SMCs in the media (Moss and Benditt 1970; Kocker et al. 1984; Glukhova et al. 1991; Frid et al. 1994). The functional significance of the expression of PDGF-C by

ATII in pup SMCs versus adult SMC is still unknown; however, this result could be validated by studies on the expression of PDGF-C in atherosclerotic human arteries

(Karvinen et al. 2009). Certain genes expressed by pups WKY12-22 SMCs have been shown to localize to the rat neointima after balloon catheter denudation (Giachelli et al.

1993). Consequently, these pup SMCs may respond better to stimuli that make them proliferate and migrate to the intima, compared to adult SMCs. PDGF-B mRNA expression is characteristic of pup SMCs and the neointima in vivo (Majesky et al. 1988)

After balloon catheter injury, PDGF-B synthesized by neointimal SMC stimulates media

SMC migration (Jawien et al. 1992). Hence, based on the notion of SMC heterogeneity, it could be accepted that if cells which resemble the “synthetic phenotype” or pups, exist in human arteries, it may be speculated that these cells proliferate and migrate in response to pathological stimuli, contributing to plaque formation. Work by Giachelli et al. (Giachelli et al. 1993) have validated this hypothesis by demonstrating that SMCs in plaque but not in the media express osteoponin, another gene expressed by pup WKY12-22 SMCs. 148

By providing evidence for a novel ATII-AT1R-Egr-1-PDGF-C axis in SMCs with a

“synthetic” phenotype, but not in “contractile” SMCs, where ATII clearly induces Egr-1 but not PDGF-C. Importantly, the induction of PDGF-C by ATII in SMCs from neonatal origin but not in adults, opens a new door for investigating how phenotypic differences between SMC subtypes affect the transcriptional regulation of PDGF-C. This information will be of great importance as it provides a more detailed understanding of SMCs heterogeneity.

The limitation of this study lies in the fact that it fails to directly determine the mechanism

(s) through which PDGF-C associates with atherosclerosis. It would be important to explore the effect of PDGF-C in the response of SMC to injury, particularly to determine a role in SMC migration. These in vitro methods using isolated cells with a culture environment that is specified and controlled, facilitates the interpretation of biochemical processes which drive SMC to respond to ATII and activate PDGF-C. However, these techniques cannot fully represent the physiological in vivo human or animal response. For instance, WKY12-22 SMCs were exposed to a continuous concentration of ATII (10-7 M), it would be difficult to relate this constant exposure to specific doses in animals with

PDGF-C expression in the vasculature. Another limitation of in vitro models is the addition of precise parameters, i.e. the addition of x concentration of Egr-1 antibody in

ChIP and EMSA assays, which may not represent the in vivo situation of transcriptional regulation of PDGF-C by ATII. Observations from these experiments require further experiments to validate the hypothesis and finally to be proven in in vivo experiments.

However, it must be appreciated that in vitro experiments were performed repeatedly under 149

similar conditions which best mimic the disease state. Further, the in vitro approaches used in Chapter 3, were meticulously planned and based on highly educated understanding of processes which were validated in vitro, including FGF-2-induction of Egr-1 in WYK12-22

SMCs, induction of Egr-1 by other members of the PDGF family and the evidence of the role for Egr-1 in SMC proliferation and migration. Lastly, other limitation is that as well as investigating endogenous expression of PDGFC by ATII, experiments were also performed measuring the effect of exogenous expression of PDGF-C. The limitation of exogenous is the level of transfection efficiency.

150

4 Conclusions and future directions: “Angiotensin II induction of

PDGF-C expression is mediated by angiotensin II receptor 1-

dependent Egr-1 transactivation”

The work in Chapter 3 was carried out to assess the effect of ATII on PDGF-C gene expression in SMCs. Our findings have provided the first evidence that ATII can significantly induce PDGF-C expression in SMCs, therefore defining ATII as an agonist of all four known PDGF-ligand chains. Mechanistically, the data demonstrated that ATII induction of PDGF-C in neonatal WKY12-22 SMCs occurs at the level of transcription via an ATII/AT1R/ERK1/2/Egr-1 signaling pathway, and that Egr-1 binding to the PDGF-C promoter occurs via a novel upstream element that cooperatively binds both Sp1 and Egr-1

(Figure 4.1). Intriguingly, ATII stimulation of PDGF-C appears to be SMC subtype specific with Egr-1-dependent and -independent transcriptional control of PDGF-C between rat neonatal and adult SMC subtypes.

4.1 Conclusions

4.1.1 ATII is confirmed to be an agonist of all four known PDGF ligand chains.

ATII and PDGFs play pivotal roles as atherogenic factors (Jawien et al. 1992; Griendling et al. 1997; Khachigian et al. 2000; Bonner 2004; Karvinen et al. 2009), thus our results on

ATII stimulation of PDGF-C were fascinating. ATII can stimulate transcription of all the other PDGF ligand chains PDGF-A, PDGF-B and PDGF-D in SMCs (Day et al. 1999;

Deguchi et al. 1999; Liu et al. 2006). These findings extend the previous reports on PDGF- 151

C-inducible by FGF-2. FGF-2 like ATII, promotes SMCs migration, plays a role in SMC phenotypic switching and is believed to also possess atherogenic properties (Hughes 1993)

(Refer to Chapter 1). In SMCs, ATII signaling to each PDGF ligand is however not straight forward, and appears to involve complex signaling pathways somewhat specific for each ligand. ATII induction of PDGF-D involves the transcription factors Ets and Sp1as well as the endogenous generation of H2O2 (Liu et al. 2006). ATII induction of PDGF-B is dependent upon Ras, ERK and JNK mechanisms (Deguchi et al. 1999). ATII induction of

PDGF-A is dependent upon ERK1/2/Egr-1/AT1R signaling pathway in vascular SMCs

(Day et al. 1999). IL-13-inducible PDGF-C expression has been shown to be down- regulated in Egr-1(-/-) and Stat6(-/-) lung fibroblasts as compared to wild type Egr-1(+/+) and Stat6(+/+), suggesting that this induction occurs via Egr-1 and Stat6 (Ingram et al.

2006). Consistent with this, we have established that ATII induction of PDGF-C occurs via

Egr-1, placing Egr-1 as an important transcription factor controlling PDGF-C expression upon diverse stimuli and in different cell types.

4.1.2 ATII-induction of PDGF-C is controlled by Egr-1 and occurs through a novel

Egr-1 binding site in the PDGF-C promoter requires cooperative interactions

between Egr-1 and Sp1 in neonatal WKY12-22 SMCs

Egr-1 is a master transcription factor, controlling the expression of multiple pathophysiological genes (McCafrrey et al. 2000; Khachigian 2006). Our findings demonstrating that Egr-1 signaling is involved in PDGF-C expression, confirms the role of

Egr-1 on this gene expression. Midgley et al. (Midgley and Khachigian 2004) show that

FGF-2 can up-regulate PDGF-C in SMCs via ERK1/2/Egr-1 signaling pathway. Although, 152

other studies have established that ATII up regulates PDGF ligands, to date only ATII- inducible PDGF-A expression is shown to be mediated by Egr-1 (Khachigian et al. 1995;

Day et al. 1999). Interestingly, the proximal region of the PDGF-C promoter (-35 to -1) which mediates FGF-2-inducible PDGF-C expression in SMCs via an Egr-1/Sp1 element

(Midgley and Khachigian 2004), is not required upon ATII stimulation. The basis for this difference is still unclear. In this study, a novel Egr-1 binding site in the PDGF-C promoter, located -543/-535 bp upstream of the transcriptional start site is shown to mediate

PDGF-C activation by ATII. Functional analysis of the Egr-1/Sp1 overlapping site (-

543GCC CCC GC-535 ) revealed that the integrity of this element is required for Egr-1 binding to activate the PDGF-C promoter.

Remarkably, it has demonstrated that Egr-1 can interplay with Sp1, which is common in the induction of PDGFs and other genes (Khachigian et al. 1995; Khachigian et al. 1996).

In ATII-inducible PDGF-A expression, Egr-1 competes with Sp1 for binding to the PDGF-

A promoter (Khachigian et al. 1995). Egr-1 and Sp1 are also shown to transactivate the human copper-zinc superoxide dismutase (Cu, Zn-SOD) promoter upon PMA exposure.

This activation however occurs by Sp1 and Egr-1 binding to (-71/-29) binding elements in the Cu, Zn-SOD promoter which do not have the consensus sequences for these transcription factors (Minc et al. 1999). In addition, suppression of hepatocyte growth factor (HGF) receptor by oxidative stress is believed to occur by a reduced Sp1 binding to its cognate sites, potentially by induction of Egr-1 (Zhang and Liu 2003). In contrast, another study has demonstrated that Egr-1 and Sp1 have “additive” effects on the activation of the chromogranin A promoter upon gastrin stimulation (Raychowdhury et al. 2002). 153

Consistent with these findings, in this thesis, Egr-1 and Sp1 are shown to cooperatively interact and bind to the novel (-543/-535) binding site in the PDGF-C promoter. Sp1 predominantly binds to (-543/-535) binding site under basal conditions. Upon ATII stimulation, Egr-1 binding is significantly induced, with invariable Sp1 level. The diverse effect of Egr-1 and Sp-1 in the activation of the PDGF-C promoter allows speculations on the potential regulatory role played by each transcription factor. For example, Sp1 may be in charge of controlling PDGF-C basal levels, whereas, Egr-1 may be controlling its induction following stimuli exposure, also seen in IL-13-inducible PDGF-C expression(Ingram et al. 2006). Previous reports have established that Sp1 is expressed in greater levels in the nuclei of unstimulated neonatal SMCs than in adult rat aortic SMCs

(Rafty and Khachigian 1997). Thus, this differential gene expression might also account for the SMCs subtype-dependent response observed in this work (see below).

4.1.3 ATII positively regulates PDGF-C expression in a SMC-subtype dependent

manner

WKY12-22 SMCs and WKY3M-22, correlate with the “synthetic” and “contractile” phenotypes of SMCs in atherosclerotic lesions in both humans (Glukhova et al. 1991), and animal models (Frid et al. 1994). WKY12-22 cells express mRNA for PDGF-A and -B

(Majesky et al. 1990; Lemire et al. 1994), and are deficient in PDGFR-α (Lemire et al.

1994). WKY12-22 also strongly express the ECM proteins tropoellastin, α1 procollagen

(type I) and osteoponin (Majesky et al. 1990; Lemire et al. 1994). In contrast, WKY3M-

22, poorly express PDGF (Lemire et al. 1994), strongly express PDGFR-α (Lemire et al.

1994) and versican (Lemire et al. 1996). Neonatal SMCs are more representative of the 154

proliferative VSMCs observed in atherosclerotic and neointimal hyperplasia (Lemire et al.

1994; Kavurma and Khachigian 2003). This study has demonstrated that the SMC subtypes, share similarities (Egr-1 up-regulation by ATII), but also have important differences in the response to ATII stimulation. ATII stimulation of PDGF-C is specific to

SMCs derived from the aortae of rat neonatal (WKY12-22) but not to SMCs derived from the aortae of rat adult SMCs. Interestingly, ATII-induction of PDGF-B mRNA expression also occurs in neonatal SMCs but not in adult SMCs (Deguchi et al. 1999). Differences in phenotype may account for these divergent responses to stimuli exposure. WKY12-22 cells proliferate and migrate more rapidly than WKY3M-22 cells. WKY12-22 cells have a greater proportion of cells in S-phase compared to WKY3M-22, as well as a greater

ERK1/2 activity (Kavurma and Khachigian 2003), which is consistent with the capacity of

WKY12-22 cells to produce their own growth factors (Majesky et al. 1992). Furthermore,

WKY12-22M are more responsive to stimulation by growth factors compared to WKY3M-

22(Majesky et al. 1992) .

4.1.4 ATII positively regulates PDGF-C expression through AT1R.

ATII can act through two different receptors AT1R and AT2R. AT1R is ubiquitously expressed in the cardiovascular system and mediates most of the physiological and pathophysiological actions of ATII (Lemarie and Schiffrin 2010) (Chapter 1, Figure 1.9).

In contrast, AT2R is highly expressed in the developing fetus, and lowly expressed in the cardiovascular system of the normal adult (Mehta and Griendling 2007; Lemarie and

Schiffrin 2010). Although the precise role of AT2R remains unclear, it appears to have opposite roles to that of the AT1R (Berk 2003). Inhibition of AT1R by Telmisartan or 155

Irbersartan results in reduced ATII induction of PDGF-B in retinal pericytes, as well as in hepatic stellate cells respectively (Amano et al. 2003; Li et al. 2008). Results in this thesis show that Losartan (inhibitor of AT1R) suppresses ATII-inducible PDGF-C expression in neonatal WKY12-22. The AT2R inhibitor (PD123319) did not affect this induction.

Moreover, this result builds on the long held belief that the injurious effects of ATII occurs when signals through the AT1R (Berk 2003; Masataka and Fukuda 2010). For example,

ATII-induction of PDGF-A, PDGF-B and PDGF-D, is shown to occur via the AT1R in vascular SMCs (Day et al. 1999; Deguchi et al. 1999; Liu et al. 2006); induction of these ligands is implicated in vascular pathology processes including migration of vascular

SMCs, inflammatory responses and extracellular matrix degradation (Chapter 1).

Importantly, studies by Nakajima, et al (Nakajima et al. 1995), have shown that AT2R expression in cultured vascular SMCs antagonizes the growth-promoting effects of the

AT1R and proliferation induced by PDGF. On the other hand, in vitro AT2R stimulation, inhibits vascular SMCs growth and proliferation and stimulates apoptosis (Lemarie and

Schiffrin 2010). These results are also supported by the notion that AT2R stimulation may contribute to the inhibition of restenosis following balloon angioplasty (Peters et al. 2001).

156

Figure 4.1 Angiotensin II induction of PDGF-C expression is mediated by AT1 receptor-dependent Egr-1 trans-activation. Schematic representation of the findings presented in Chapter 3. ATII signals via AT1R to activate Egr-1 transcription and binding to the PDGF-C promoter (-543/-534). This in turn activates PDGF-C transcription. ATII- induction of PDGF-C is significantly dependent on Egr-1 (as demonstrated by the use of the Egr-1 DNAzyme) and the AT1R (as demonstrated by the use of Losartan (AT1R inhibitor). Dashed line represents the “potential” pathway by which Egr-1 is activated.

157

4.2 Future directions

This is one of the few studies investigating the transcriptional activation of PDGF-C upon

ATII stimulation in SMCs. As a result of its novelty, the lack of supportive or contradictory data on molecular mechanisms controlling PDGF-C expression, makes this ligand an attractive candidate to further investigate. This research initiated new findings and expanded the current knowledge of how this relatively new PDGF ligand is regulated.

Future experiments clarifying the role of PDGF-C in SMCs are presented below.

4.2.1 Does ERK1/2 play a role in ATII-inducible PDGF-C expression?

JNK, p38 and ERK1/2 are essential for proliferation of WKY12-22 and WKY3M-22

SMCs, however, ERK1/2 signaling is required for migration of neither SMC subtype

(Kavurma and Khachigian 2003). ERK1/2 has been shown to regulate Egr-1 expression upon a variety of stimuli (Cahill et al. 1996; Deguchi et al. 1999; Gurjar et al. 2001;

Amano et al. 2003; Fahmy et al. 2003; Ingram et al. 2006) and FGF-2-inducible PDGF-C expression has been linked to ERK1/2 signaling pathway (Midgley and Khachigian 2004).

ATII can induce PDGF-A expression via the ERK1/2 signaling pathway (Day et al. 1999), however, ATII-inducible PDGF-B expression occurs via both ERK1/2 and c-Jun kinase

(JNK) (Deguchi et al. 1999). It is still unclear which signaling cascade ATII employs to induce PDGF-C expression.

Inhibitor studies could be useful in identifying how signaling pathways can affect induction of PDGF-C by ATII. This study will involve the use of a range of pharmacological 158

inhibitors to a variety of signalling pathways (PD98059, inhibitor of ERK1/2; SB203580, inhibitor of p38, SP600125, JNK inhibitor) followed by ATII-stimulation in neonatal

WKY12-22 SMCs. These studies would increase our understanding on the signaling pathways associated with PDGF-C gene expression.

4.2.2 Is Egr-1 and PDGF-C involved in ATII-induced migration of vascular SMCs?

Proliferation and migration of SMCs are key cellular processes underlying intimal thickening and progression of atherosclerosis (Ross 1986; Kavurma and Khachigian 2003).

We demonstrated that ATII induces Egr-1 at the level of mRNA and protein in neonatal

WKY12-22 SMCs. Furthermore, ATII is a chemo attractant for SMCs (Saito et al. 2002;

Mehta and Griendling 2007; Masataka and Fukuda 2010). However, whether Egr-1 or

PDGF-C could modulate ATII neonatal WKY12-22 SMCs migration and proliferation needs to be established. To address this, the in vitro wound assay could be utilized. Knock down/silencing technology (e.g. Egr-1 DNAzyme and PDGF-C siRNA) as well as neutralising antibodies for these genes would confirm whether Egr-1and PDGF-C play a role in this process.

4.2.3 The role of Sp1 on the PDGF-C promoter activation by ATII

Results presented in this study demonstrate that Egr-1 cooperatively associates with Sp1 to activate the PDGF-C promoter when stimulated with ATII. We also showed that when the

-543/-535 response element in the PDGF-C promoter is mutated, the PDGF-C promoter is not activated by Egr-1. In contrast, Sp1 is able to activate the promoter. Although Sp1 159

binds to this element, Sp1 transactivation of the PDGF-C promoter might occur via another binding site in the PDGF-C promoter.

Studies have demonstrated that Sp1 phosphorylation potentially modulates transcriptional activation of diverse genes including PDGF-B (Rafty and Khachigian 2001), PDGF-D (Tan et al. 2008) and MT1-MMP (Yun et al. 2002). Sp1 could be phosphorylated by many kinases, particularly PKC. Thus, studies on phosphorylation status of Sp1 upon ATII might provide new signalling pathways by which ATII induces PDGF-C expression. Recently, we have demonstrated that following mechanical injury, Sp1 is phosphorylated (Thr453).

This phosphorylation increases the physical interaction between p-Sp1(Thr453) and NF-κB in the TRAIL promoter and promotes VSMC proliferation (Chan et al. 2010).

Furthermore, studies by Tan et al (Tan et al. 2008), using novel antibodies recognizing the

PKC-zeta-phosphorylated form of Sp1, have shown that ATII stimulates phosphorylation of PKC-zeta, which in turn phosphorylates three amino acids (Thr668, Ser670, and Thr681) in the zinc finger domain of Sp1. Interestingly, all three residues were required for Sp1- dependent PDGF-D activation in response to ATII.

Thus, it would be interesting to determine whether phospho-Sp1 controls PDGF-C regulation. The use of this novel Sp1 antibody (p681) will expand the limited knowledge of the role of Sp1 in the PDGF-C expression. PDGF-C expression has been localised to macrophages, SMCs and EC (Wagsater et al. 2009). It would be interesting to determine whether Egr-1 and phospho-Sp1 also co-expressed in region of the plaque expressing 160

PDGF-C. These experiments may provide new insights on the regulatory roles of Sp1 on the PDGF-C expression as well as on the expression and localization pattern of PDGF-C,

Egr-1 and Sp-1 in human atherosclerotic plaques.

4.2.4 Which other transcription factors mediate ATII-inducible PDGF-C

expression?

We have identified Egr-1 as the transcription factor essential for the induction of PDGF-C by ATII in SMCs. It would be particularly interesting to determine which other transcription factor(s) may control PDGF-C expression in response to ATII exposure.

“MatInspector” (Genomatix, USA), can identify potential transcription binding elements within the given nucleotide sequences. This software can be utilised to identify transcription factor binding sites in the PDGF-C promoter. For instance, PDGF-C contains putative binding elements for Ets-1 and Smad (Figure 4.2). Once putative regulatory sites have been identified, EMSA, ChIP and mutational analyses will confirm the functionality of these sites. Thus, identification of new regulatory elements will expand the “limited” knowledge on the transcriptional activation of this relative new member of the PDGF family.

4.2.5 Explore phenotypical differences between SMCs subtypes

SMCs subtypes are known to share similarities and differences in signalling pathways

(Kavurma and Khachigian 2003), which might affect expression of particular genes relevant to each SMC subtype. For instance, SMCs in middle/advanced-stage 161

atherosclerotic lesions, express fibroblast surface protein (FSP), whereas, FSP is not expressed in early lesions (Martinez-Gonzalez et al. 2002). Additionally, WKY12-22

SMCs have higher levels of phosphorylated Elk-1 in comparison to WKY3M-22 SMCs

(Kavurma and Khachigian 2003). Elk-1 is downstream of ERK1/2 and can bind to cis- acting elements to activate the Egr-1 promoter upon IL-1beta stimulation (Wang et al.

2010). The novel findings presented in this study demonstrates that although ATII up- regulates Egr-1 expression in both neonatal WKY12-22 and adult WKY3M-22 SMCs,

PDGF-C is only induced in SMCs from neonatal origin. Thus, experiments, which might elucidate this divergent response, could involve assessment of expression profiles of ATII receptors in pups SMC. To further elucidate ATII signaling in WKY122-22 SMCs, western blot experiments comparing expression levels of AT1R and AT2 would validate the role of AT1R in ATII-inducible PDGF-C expression. Total protein extracts of quiescent and ATII-stimulated WKY12-22 SMCs will be subjected to western blot analysis to determine the expression levels of the ATII receptors. as well as EMSA and ChIP analyses to investigate the binding profile of Egr-1 to the PDGF-C promoter in response to

ATII. Given these results, the use of other primary cell lines (i.e. human aortic SMCs) which resembles an in vitro model of the “contractile phenotype”, would validate and expand the phenotype-dependent” findings by PDGF-C expression in SMC”. “

4.2.6 Role of PDGF-C in atherogenesis

As far as the role of PDGF-C in atherosclerosis is concerned, no in vivo studies have been performed to demonstrate its role in neointima formation. However, based on our findings, 162

particularly the SMC phenotype-dependent response to ATII, the role of PDGF-C in atherogenesis could begin to be studied.

Rat ballon catheter experiments, a well-established animal model of vascular SMC injury, could be used to validate the notion of PDGF-C expression in neointimal SMCs (which are represented by pups WKY12-22 SMCs used in in vitro experiments) by ATII. Balloon injury to rat carotid arteries results in instant denudation of the endothelium, SMC hyperplasia is observed within 4 days and neoitima formation occurs 14 days after (Clowes et al. 1983). The experiment will examine the role of PDGF-C with or without the infusion of ATII together with or without the delivery of PDGF-CC neutralizing antibodies (this approach has been used in studies demonstrating that blockade of PDGF-CC by a single dose of 2ug of antibody per eye inhibits choroidal angiogenesis (Hou et al. 2010). Injured arteries will be collected over a time course (i.e. 2, 8, 12, 24h and then daily for 14 days), homogenized and used in PCR experiments to determine the optimal time for the determination of PDGF-C in the tissue. In another group of animals, carotid arteries will be perfusion fixed with formalin and used for (a) IHC analysis for PDGF-C expression. This will prove the spatial localization of PDGF-C in SMCs of injured vessels and not in SMCs of intact vessels (which represent the adult phenotype used in the in vitro experiments). (b)

IHC of carotid artery sections to demonstrate the effect of PDGF-C neutralizing antibodies on neointimal hyperplasia, with the ultimate goal of establishing the role of PDGF-C in atherogenesis.

163

Figure 4.2 Proximal region of the PDGF-C promoter. Putative binding elements for

Sp1 (green), Egr-1(Black et al.), Smad (red) and Ets (blue) in the PDGF-C promoter. 164

5 Egr-1 induction by PDGF-BB involves MMP/ADAM-dependent

ERK1/2, PDGFRβ and EGFR phosphorylation and PDGFRβ-EGFR

heterodimerisation

5.1 Introduction

Intimal hyperplasia is a key feature in the development of atherosclerosis and also restenosis following percutaneous coronary intervention (Khachigian 2006). SMC hyperplasia and extracellular matrix deposition contribute to neointima formation

(Schwartz 1997). Intimal thickening is further exacerbated by infiltration of inflammatory cells (T-cells, monocytes). Injury, inflammation and wall stress are potentially the major stimuli for intimal hyperplasia (Newby 2000).

Heterogeneity of the SMC population is another characteristic in the vessel wall (Owens et al. 2004). In adult blood vessels, contractile SMCs have limited proliferative and migratory capabilities which might be restricted to their contractile functions (Kavurma and

Khachigian 2003; Owens et al. 2004). Nevertheless, in response to insult or injury, these

SMCs undergo transient phenotypic variation to a highly migratory and proliferative state, the synthetic phenotype (Newby and Zaltsman 2000; Owens et al. 2004). This phenotypic modulation triggered by changes in local environmental cues (growth factors, mechanical forces, cell-cell and cell-matrix interactions and inflammatory factors) is driven by an altered gene expression of pathophysiologically relevant genes. 165

Among the immediate-early genes rapidly induced in the injured artery wall is the zinc finger transcription factor, early growth response-1 (Egr-1) (McCafrrey et al. 2000). Egr-1 is transiently induced by growth factors including platelet-derived growth factor (PDGF), cytokines, hypoxia, oxidised lipoproteins, shear stress, angiotensin II (ATII), and other injurious stimuli (Khachigian 2006). Egr-1 is poorly expressed in the normal artery wall, but is expressed in response to injury and mediates a series of transcriptional changes that lead to the altered expression of genes such as PDGF, transforming growth factor-β1

(TGFβ-1), MMPs, tissue factor (TF) (Khachigian et al. 1996). Egr-1 transcription is dependent on RAS-RAF-MEK-ERK1/2 signalling and serum response elements in the Egr-

1 promoter (Wang et al. 2006). Egr-1 plays a vital role as a mediator of SMC growth and intimal thickening as part of the restorative response to vascular injury. For example, inhibition of Egr-1 by catalytic DNA molecules blocks SMC replication and regrowth after in vitro scraping injury and intimal thickening after balloon injury in rats (Santiago et al.

1999), or after permanent ligation in rat carotid arteries (Lowe et al. 2002). Similar strategies targeting Egr-1 in pig coronary arteries reduce in-stent restenosis (Lowe et al.

2001). Egr-1 is expressed in human and animal models of atherosclerosis (McCafrrey et al.

2000).

Egr-1 is induced by numerous stimuli (Santiago et al. 1999; Khachigian 2004; Midgley and

Khachigian 2004). Of particular interest is the induction of Egr-1 by PDGF-BB in SMCs

(Kamikura et al. 2004). PDGF-BB is associated with atherosclerosis due to its mitogenic effect on SMCs making them to proliferate and migrate from the media into the intima

(Jawien et al. 1992). PDGF-B mRNA and protein is present in human atherosclerotic tissue 166

and in human coronary arteries after percutaneous transluminal coronary angioplasty (Ross et al. 1990). Infusion of recombinant PDGF-BB stimulates SMC migration and intimal thickening (Jawien et al. 1992). Furthermore, PDGF-B gene transfer into pig arteries stimulates SMC proliferation and intimal hyperplasia (Nabel et al. 1993; Pompili et al.

1995).

In addition to Egr-1, extracellular proteases such as plasminogen activators and MMPs are induced during vascular injury. These proteases contribute to neointima formation and indeed, plaque instability, by degrading matrix and non-matrix substrates (Newby 2006).

MMPs production is regulated by growth factors and cytokines (Chang and Werb 2001;

Newby 2005). Active MMPs are produced from pro-MMP forms by the local action of proteases (Chan et al. 2006) Once activated, MMPs influence a diverse range of cellular mechanisms including cell proliferation, migration, matrix remodelling (Mifune et al.

2005). MMPs cleave latent growth factors, whereby the cleaved active ligand in turn binds and activates its receptor (Chang and Werb 2001). A prototypic example of growth factor activation by MMP-dependent shedding is epidermal growth factor receptor (EGFR) activation (Mifune et al. 2005). This proposed model of EGFR transactivation is described as “the triple membrane-passing signalling” (Prenzel et al. 1999), whereby activation of a

G-protein-coupled receptor (GPCR), involves MMP-dependent shedding of pro-heparin binding epidermal growth factor like (HB-EGF) and its binding to EGFR. (Chapter 1,

Figure 1.16).

167

There is a growing indication that other than trans-activation by the GPCR-

MMPs/ADAMs-ligand release and binding, EGFR is also transactivated by direct induction of EGFR tyrosine kinase activity (Zhou et al. 2007). PDGF has been shown to transactivate EGFR (Habib et al. 1986; Saito and Berk 2001; Mendelson et al. 2010), however, the mechanisms for PDGF-inducible EGFR transaactivation remain to be clarified. To date, it has been shown that PDGF-stimulated migration of murine fibroblasts is dependent on EGFR expression and tyrosine phosphorylation (Li et al. 2000). Saito et al. (Saito et al. 2001), have also established that PDGFR and EGFR heterodimerise, basally and more so upon PDGF-BB exposure. The receptor heterodimerisation is believed to be

Src or ROS-dependent. Moreover, disruption of this receptor association results in inhibition of PDGF-BB-inducible ERK1/2 activation in SMCs.

The aim of this study was to explore the mechanisms by which PDGF-BB induces Egr-1 expression in vascular SMCs. Specifically, we examined the role of MMPs/ADAMs on

PDGF-BB-inducible Egr-1 expression. Pharmacologic inhibitors which inhibit the activity of MMPs and ADAMs were utilised. In parallel, we investigated the effect of stimulating the activity of MMPs in vitro. The nature of the results, and the reported evidence regarding EGFR transactivation by PDGF-BB, prompted us to investigate the role of EGFR in PDGF-BB-induced Egr-1 expression. The surprising ability of MMP/ADAM inhibitors to down-regulate PDGF-BB-inducible Egr1 expression and the results confirming the vital role of EGFR on PDGFR-signalling, led to investigations into the transduction pathway upstream of Egr-1 in the context of PDGFRβ-EGFR heterodimerisation. This chapter 168

therefore interrogated the role of MMP/ADAM in the induction of Egr-1 and the involvement of receptor crosstalk in vascular SMCs exposed to PDGF-BB.

5.2 Material and methods

5.2.1 Chemicals

A comprehensive list of pharmacologic inhibitors, final concentration and incubation times is shown in Table 4.1.

5.2.2 Cell culture

WKY12-22, RASMCs and HASMCs were maintained as described in Chapter 2, Section

2.2.

5.2.3 Total RNA preparation and cDNA synthesis

SMCs were grown and-serum arrested for 24 h. SMCs were exposed to PDGF-BB (50 ng/ml) for various times. In inhibitor studies, prior to PDGF-BB stimulation, SMCs were pre-treated with pharmacologic inhibitors at concentrations and for times listed in Table

5.1. PDGF-BB stimulation was terminated by addition of 1 x ice-cold PBS and RNA extraction and cDNA synthesis carried out as described in Chapter 2 (Sections 2.3.1 and

2.3.2)

169

Table 5.1 List of inhibitors used. (£) Chemical dissolved in dH2O. (*) Reagents dissolved in DMSO.

Inhibitor Targeted Gene Final Pre- Source (Cat /Activator Concentration Incubation number)

AG825 EGFR type II 10 μM (*) 30 min Calbiochem (121655)

AG1295 PDGFRβ 5 μM (*) 30 min Calbiochem (658550)

A 1478 EGFR 5 μM (*) 30 min Calbiochem (658552)

Amiloride u-PA 50 μM, 100 μM 30 min Sigma (A7410) (*)

Amino Caproic Plasmin 50 μM, 100 μM 30 min Sigma (A2504) Acid (£)

APMA Metalloproteinase 30 μM (*) 1 h Calbiochem Activator (164610)

BiPS MMP-2/MMP-9 10 μM (*) 30 min Calbiochem inhibitor II (444249)

CRM197 HB-EGF blocker 10 ng/ml (£) 30 min Sigma (B0564)

GM6001 Broad spectrum 10 μM (*) (IP 30 min Calbiochem MMP inhibitor experiments) 25 (364205) (MMP-1,-2,-3,-8 μM (mRNA and and -9) Western Blotting)

GM6001 Inactive analogue 10 μM (*) (IP 30 min Calbiochem (Negative of GM6001 experiments) 25 (364210) Control) μM (mRNA and Western Blotting)

Losartan ATII Receptor 1 1 μM (£) 1 h Gift from (AT1) Merck, Sharp & Dohme

MMP-2 Highly selective 10 μM (*) 30 min Calbiochem inhibitor of (444244) MMP-2 170

MMP-3 Highly selective 10 μM (*) 30 min Calbiochem inhibitor of (444218) MMP-3

MMP-9 Highly selective 10 μM (*) 30 min Calbiochem inhibitor of (444278) MMP-9

PD123319 ATII Receptor 2 10 μM (*) 30 m Sigma (P186) (AT2)

PD153035 EGFR 5 μM (*) 30 min Calbiochem (234490)

PD98059 ERK1/2 30 μM (*) 1 h Sigma (P215)

SB203580 p38 MAPK 10 μM (*) 1 h Sigma (P4558270)

SP600125 JNK 10 μM (*) 1 h Sigma (S5567)

TAPI-1 ADAM-17 10 μM (*) 30 min Calbiochem (579051)

U0126 ERK 10 μM (*) 30 min Calbiochem (662005)

171

5.2.4 Real-time PCR

PCR amplification was performed to evaluate the mRNA expression of Egr-1, MMP-2,

MMP-9, ADAM17 and β-actin as described in Chapter 2 (Section 2.3.4). Primer details and amplification conditions are in Table 5.2.

5.2.5 Co-immunoprecipitation

A total of 500 μg of protein extracts from serum-arrested WKY12-22 SMCs exposed to

PDGF-BB for 0, 5, 10 and 20 min or treated with MMP/ADAM inhibitors prior the addition of PDGF-BB were subjected to co-immunoprecipitation experiments utilising a Reversible

Catch and Release Immunoprecipitation System ® v2.0 (Millipore Corporation, USA). The method was carried out according to manufacturer’s instructions. Briefly, Spin columns were washed twice with 400 μL of 1 x Catch and Release wash buffer by centrifugation at

5000 rpm for 30 s before pipetting the reagents in the following order: 500 μg of whole cell lysates, 10 μL of PDGFRβ antibody, 10 μL of antibody capture Affinity ligand and 1 x wash buffer to a final volume of 500 μL. Columns were incubated overnight at 4 ºC with gentle spinning ensuring that the slurry was constantly mixed during incubation. Spin columns were then washed three times with 400 μL of 1 x wash buffer by centrifugation at

5000 rpm for 30 s. Columns were placed into new capture eppendorf tubes. Proteins were eluted with 70 μL of 1 x denaturing Elution buffer containing β-mercaptoethanol to a final concentration of 5 % (v/v). Saved eluates were boiled in SDS buffer for analysis by

Western blotting (Sections 2.4 and 5.2.6). Proteins were resolved on a 6 % SDS-PAGE gel.

172

Table 5.2 Real-time PCR conditions and primer sequences

Real Time PCR conditions for ALL the genes were as followed:

Initial Hold Tº: 50 ºC for 2 min

Hold Tº: 94 ºC for 10 min

Cycling conditions: 94 ºC for 20 s, 60 ºC for 45 s, 72 ºC for 20 s

Rat Egr- 1 (Cycle number: 25 x) Product size: 230 bp

Forward sequence: 5’-GCC TTT TGC CTG TGA CAT TT-3’

Reverse sequence: 5’-AGC CCG GAG AGG AGT AAG AG-3’

Rat Beta-Actin (Cycle number: 25 x) Product size: 228 bp

Forward sequence: 5’-AGC CAT GTA CGT AGC CAT CC-3’

Reverse sequence: 5’-CTC TCA GCT GTG GTG GTG AA-3’

Rat MMP-2 (Cycle number: 40 x) Product size: 230 bp

Forward sequence: 5’-GAT ACC CTC AAG AAG ATG CAG AAG T -3’

Reverse sequence: 5’-ATC TTG GCT TCC GCA TGG T-3’

Rat MMP-9 (Cycle number: 40 x) Product size: 230 bp

Forward sequence: 5’-TCA AGG ACG GTC GGT ATT GG-3’

Reverse sequence: 5’-ACG TGC GGG CAA TAA GAA AG-3’

Rat ADAM-17 (Cycle number: 40 x) Product size: 230 bp

Forward sequence: 5’-ACG TAA TTG AGC GGT TTT G-3’

Reverse sequence: 5’-TCC ATG CTG CTC AAC ATC TC-3’

173

5.2.6 Western blot analysis

SMCs were plated onto 100-mm2 petri dishes and grown to 70 % confluency in 10 mL of

10 % FCS-WM. SMCs were serum starved for 24 h. SMCs were exposed to PDGF-BB

(50 ng/mL) for 0.25, 0.5, 1, 2 and 4 h or were treated with MMP/ADAM inhibitors for 30 min prior to the addition of PDGF-BB for 30 min. SMCs were washed with 1 x cold PBS to terminate the specified treatments. Western blot analysis was performed as described in

Chapter 2 (Section 2.4). Proteins were resolved on a 6 % SDS-PAGE resolving gel.

Incubation with primary antibodies was performed as detailed in Table 5.3. Incubation with the respective secondary antibody was for 1 h at a dilution of 1:1000 in 5 % skim milk powder in 0.05 % (v/v) PBS-T at RT for all antibodies. Chemilumescence was detected using HRP-linked reagents (Section 2.4).

174

Table 5.3 Antibodies used in Western blotting and Co-IP experiments. (a) Dilutions were performed in 0.05 % PBS-T. SM=Skim milk

Primary Type Dilution Factor (a) Incubation Company

Antibody (Resuspended in)

β-actin Mouse 1:30 000 1 h, RT Sigma

Monoclonal 5 % (w/v) SM (A5316)

Egr-1 Rabbit 1:1000 O/N, 4 ºC Santa Cruz

Polyclonal 1 % (w/v) BSA (Sc-189)

p-ERK Rabbit 1:1000 1 h, RT Santa Cruz

Polyclonal 1 % BSA (w/v) Sc-7383

Total ERK Rabbit 1:1000 1 h, RT Santa Cruz

Polyclonal 1 % (w/v) BSA Sc-154

EGFR (WB) Rabbit 1:1000 O/N, 4 ºC Santa Cruz

Polyclonal 1 % (w/v) BSA Sc-03

EGFR (Co-IP) Rabbit 1:500 O/N, 4 ºC Cell Signalling

Monoclonal 5 % (w/v) BSA C-74B9

PDGFRβ (WB) Rabbit 1:1000 O/N, 4 ºC Santa Cruz

Polyclonal 1 % (w/v) BSA Sc-339

PDGFRβ (Co- Mouse 1:500 O/N, 4 ºC Cell Signalling IP) Monoclonal 5 % (w/v) BSA 2B3

p-EGFR Mouse 1:1000 O/N, 4 ºC Santa Cruz

Tyr1068 Monoclonal 1 % (w/v) BSA Sc-81488

175

5.2.7 Small interfering RNA (siRNA) studies

Growth-quiescent SMCs plated in 100-mm2 petri dishes were transfected using FuGENE6 with 100 nM siRNA targeting rat EGFR (siRNA-ON-TARGETplus, Dharmacon RNAi

Technologies, USA) or with 100 nM non-specific siRNA and incubated for 18 h at 37 ºC, prior to stimulation with 50 ng/mL PDGF-BB for 30 min. Whole cell protein lysates were collected and used in western blot as described in Section 2.4.

5.2.8 Wound scratch assay

WKY12-22 SMCs were grown in 6-well plates to 60 % confluence in 10 % FCS-WM. The cells were rendered quiescent by culturing in SFM-WM for 24 h. Cell monolayers were scratched using a P20 micropipette tip. Medium was replaced to remove cell debris.

Inhibitors were added to the cells for 30 min followed by exposure to PDGF-BB (50 ng/mL). Cell growth in the denuded zone was monitored and images at 10 x magnification were taken at 24 and 48 h after injury.

5.2.9 Statistical and densitometry analyses

Statistical and Densitometry Analyses were performed as described in Chapter 2, Sections

2.5 and 2.6 respectively.

176

5.3 Results

5.3.1 PDGF-BB transiently induces Egr-1 expression in WKY12-22 SMCs and

occurs via the ERK1/2 pathway

Previous studies from our laboratory on the zinc finger transcription factor Egr-1 suggest that it plays a key regulatory role in the induction of many genes implicated in diverse vascular pathological processes (Day et al. 1999). Induction of Egr-1 by stimuli such as growth factors (PDGF) is a rapid and transient process. We have confirmed that PDGF-BB increases Egr-1 mRNA (Figure 5.1A) and protein expression (Figure 5.1B and C) as early as 30 min in rat neonatal WKY12-22 SMCs. This induction, as expected, is mediated via

ERK1/2 since the ERK inhibitor PD98059 blocks this induction (Kamikura et al. 2004).

Activation of ERK1/2 by PDGF-BB is once again confirmed to occur upstream of Egr-1

(Figures 5.2A- C).

177

B C

Figure 5.1 PDGF-BB transiently and rapidly induces Egr-1 expression. Rat neonatal

WKY12-22 SMCs, serum-arrested for 24 h, were incubated with PDGF-BB (50 ng/mL) for various times. (A) Real time PCR showing Egr-1 mRNA expression relative to β-actin mRNA. (B) Western Blot analysis comparing cells treated with PDGF-BB at various times. β-actin blots demonstrate unbiased loading. (C) Densitometry analysis of Egr-1 relative to β-actin. * denotes p<0.05, *** p<0.001. The results are representative of at least two independent observations. Error bars represent the mean ±SE. 178

B C

Figure 5.2 PDGF-BB induction of Egr-1 occurs through ERK1/2. (A) Effect of MAP kinase inhibitors on PDGF-BB-inducible Egr-1 mRNA expression. Serum starved

WKY12-22 SMCs were pretreated with 30 μM PD98059, 10 μM SB203580 or 20 μM

SB600125 for 1 h, followed by stimulation with PDGF-BB (50 ng/mL) for 30 min. (B)

Serum-arrested SMCs were exposed to PDGF-BB for various times. Phosphorylation of

ERK1/2 in the cell lysates was estimated by Western blotting. Total ERK levels show equal loading. (C) ERK1/2 phosphorylation relative to total ERK1/2 by scanning densitometry. * denotes p<0.05, ** p<0.01, *** p<0.001. 179

5.3.2 PDGF-BB induction of Egr-1 expression in WKY12-22 SMCs is MMP/ADAM-

dependent

Activation of MMPs by GPCR with a subsequent transactivation of the EGFR, leads to the initiation of intracellular signals (i.e. ERK1/2), ultimately inducing cellular growth and tissue remodelling (Page-McCaw et al. 2007). Importantly, growth factors such as PDGF-

BB have been linked with the regulation of MMPs production (Reuben and Cheung 2006).

Whether MMPs are involved in the inducible expression of Egr-1 by PDGF-BB has not been demonstrated. To examine the involvement of MMPs in PDGF-BB-inducible Egr-1 expression, WKY12-22 SMCs were pretreated (for 30 min) with a series of pharmacological MMP/ADAM inhibitors, including GM6001 (a pan spectrum MMP inhibitor), BiPS (an N-phenylsulfonyl-hydroxamic acid derivative known as MMP-2/9 inhibitor) and TAPI-1 (an ADAM17-specific inhibitor), followed by stimulation with

PDGF-BB for 30 min. Pre-treatment of SMCs with GM6001 (Figure 5.3A-C), BiPS

(Figure 5.4 A-C) and TAPI-1 (Figure 5.5A-C) resulted in dose-dependent inhibition of

Egr-1 mRNA expression and also affected protein levels. Western blot analysis revealed that these MMP/ADAM inhibitors also blocked ERK1/2 activation (Figures 5.3B and D,

5.4B and D, and 5.5B to D) the MAPK upstream of Egr-1.

180

B C D

181

Figure 5.3 GM6001 inhibits PDGF-BB-inducible Egr-1 mRNA and protein expression in SMCs. Serum-arrested rat neonatal WKY12-22 SMCs were pre-treated with GM6001 for 30 min prior to stimulation with PDGF-BB (50 ng/mL) for 30 min. (A) Real-time PCR analysis showed the effect of GM6001 on PDGF-BB-inducible Egr-1 mRNA. DMSO was used as a carrier. The inactive analogue of GM6001, was used as a negative control. Data were normalised to β-actin. (B) Western Blotting analysis of rat neonatal WKY12-22

SMCs pre-incubated with GM6001 (25 μM) for 30 min prior to addition of PDGF-BB (50 ng/mL) for 30 min. Proteins were run on a 10 % SDS-PAGE gel and membranes were blotted with Egr-1, p-ERK1/2, total ERK1/2 and β-actin. Assessment of band intensities of

(relative to total ERK1/2) by scanning densitometry. The results are representative of at least two independent observations. *p<0.05, ** p<0.01, *** p <0.001. Error bars represent the mean ±SE.

182

B C D

183

Figure 5.4 BiPS inhibits PDGF-BB-inducible Egr-1 mRNA and protein expression in

SMCs. Serum-arrested rat neonatal WKY12-22 SMCs were pre-treated with BiPS for 30 min prior to stimulation with PDGF-BB (50 ng/mL) for 30 min. (A) Real-time PCR analysis showed the effect of BiPS on PDGF-BB-inducible Egr-1 mRNA. DMSO was used as a carrier. Data were normalised to β-actin. (B) Effect of BiPS (10 μM) on Egr-1 protein expression and ERK1/2 phosphorylation. Proteins were run on a 10 % SDS-PAGE gel and membranes were blotted with Egr-1, p-ERK1/2, total ERK1/2 and β-actin. (C)

Assessment of band intensities of PDGF-BB-inducible Egr-1 (relative to β-actin) and (D) phosphorylation of ERK1/2 (relative to total ERK1/2) by scanning densitometry. The results are representative of at least two independent observations. * denotes p<0.05; ** p<0.01; *** p<0.001. Error bars represent the mean ±SE.

184

B C D

185

Figure 5.5 Effect of TAPI-1 on PDGF-BB-induced Egr-1 expression. (A) Egr-1 mRNA levels in TAPI-1-treated rat neonatal WKY12-22 SMCs (different concentrations) for 30 min prior to PDGF-BB stimulation for 30 min (50 ng/mL). (B) Egr-1 protein and ERK1/2 phosphorylation status assessed by Western Blot when TAPI-1 (20 μM) was utilised.

Proteins were run on a 10 % SDS-PAGE gel. Assessment of band intensities of (C) PDGF-

BB-inducible Egr-1 (relative to β-actin) and (D) phosphorylation of ERK1/2 (relative to total ERK 1/2) by scanning densitometry. The data are representative of three different experiments. * denotes p<0.05,** p<0.01, *** p<0.001. Error bars represent the mean

±SE.

186

To confirm that MMP inhibitors did not affect MMP mRNA expression, cDNA from previous experiments (Figure 5.3, 5.4, 5.5) was utilised to assess MMP-2, MMP-9 and

ADAM-17 mRNA expression by real-time PCR analysis. GM6001, BiPS and TAPI-1 did not inhibit mRNA levels for MMP-2 (Figure 5.6A), MMP-9 (Figure 5.6B), or ADAM17

(Figure 5.6C). These results show that BiPS, GM6001 and TAPI-1 inhibition of PDGF-

BB-inducible Egr-1 expression was not a consequence of inhibition of MMP-2/-9 or

ADAM17 mRNA expression.

187

188

Figure 5.6 Effect of MMP inhibitors on MMP/ADAM mRNA. Quiescent rat neonatal

WKY12-22 SMCs were incubated with BiPS (10 μM), GM 6001 (GM+) (25 μM),

GM6001 inactive analogue (GM-) (25 μM) and TAPI-1 (20 μM) for 30 min, prior to

PDGF-BB exposure (50 ng/mL) for 30 min and RNA extracted (Section 2.2.1). Real-time

PCR (as described in Section 4.2.4) denotes the effect of MMP/ADAM on (A) MMP-2, (B)

MMP-9 and (C) ADAM17 mRNA. The data are representative of three different experiments. Error bars represent the mean ±SE.

189

5.3.3 PDGF-BB induction of Egr-1 expression in WKY12-22 SMCs may require

other MMPs different from MMP-2, -3 and -9

To establish the MMP(s) implicated in PDGF-BB-inducible Egr-1 expression, serum- deprived rat neonatal WKY12-22 SMCs were separately pre-treated (for 30 min) with specific inhibitors to MMP-2 (Figure 5.7A), MMP-9 (Figure 5.7A) and MMP-3, (Figure

5.7B) followed by addition of PDGF-BB (50 ng/mL). Real-time PCR analysis demonstrated that PDGF-BB-induced Egr-1 mRNA expression was not altered by these inhibitors.

190

191

Figure 5.7 Effect of inhibitors to MMP-2, MMP-3 and MMP-9 on Egr-1 mRNA.

Quiescent rat neonatal WKY12-22 SMCs were incubated with MMP-2 (MMP-2 inhibitor I,

Calbiochem, 10-25 μM), MMP-3 (MMP-3 inhibitor II, Calbiochem, 5-20 μM) and MMP-9

(MMP-9 inhibitor I, Calbiochem, 10-25 μM) for 30 min prior to PDGF-BB exposure (50 ng/mL) for 30 min and RNA extracted (Section 5.2.3). Real-time PCR (as described in section 5.2.4) shows the effect of inhibitors to MMP-2 (A), MMP-9 (A) and MMP-3 (B) on

Egr-1 mRNA expression. *** denotes p<0.001. The data are representative of three different experiments. Error bars represent the mean ±SE.

192

5.3.4 Activation of MMPs with APMA, but not with plasmin nor urokinase-type

plasminogen activator enhances PDGF-BB-induced Egr-1 expression in

WKY12-22 SMCs

In order to obtain complementary data, we evaluated the effect of APMA, a well-known in vitro MMP activator, on PDGF-BB-inducible Egr-1 expression. Pre-treatment of rat neonatal WKY12-22 SMCs cells with APMA (30 μM) for 1 h prior to incubation with

PDGF-BB (50 ng/mL), enhanced Egr-1 mRNA expression in a dose-dependent manner, compared with cells only stimulated with PDGF-BB (Figure 5.8A). APMA treatment also induced basal Egr-1 mRNA expression, with levels comparable to those induced by PDGF-

BB (Figure 5.8A). APMA not only enhanced PDGF-BB-stimulated Egr-1 protein levels

(Figure 5.8B-C), it also enhanced ERK1/2 phosphorylation (Figure 5.8B, D). As expected, the ERK1/2 inhibitor PD98059 blocked PDGF-BB’s induction of both ERK1/2 and Egr-1 (Figure 5.8B-D).

Proteases such as urokinase-type plasminogen activator (Pisarev et al.) and plasmin have also been shown to activate MMPs (Hurtado et al. 2007), but whether they play a role in

PDGF-BB-inducible Egr-1 expression has not been described. Real-time PCR analysis established that neither Amiloride nor Aminocaproic Acid, inhibitors of uPA and plasmin respectively, affected Egr-1 levels. These findings suggest that uPA and plasmin do not play a role in this signalling pathway, under these experimental conditions (Figure 5.8E-

F).

193

B C D

194

195

Figure 5.8 APMA potentiates PDGF-BB-inducible Egr-1 expression and ERK1/2 phosphorylation. Quiescent rat neonatal WKY12-22 SMCs were treated with increasing concentrations of the MMP activator APMA or vehicle for 1 h before stimulation with

PDGF-BB (50 ng/mL) for 30 min. (A) Real-time-PCR showed the effect of APMA on

Egr-1 mRNA. (B) Western blotting on APMA-treated cell extracts looking at Egr-1 protein expression and phosphorylation status of ERK1/2. PD98059 (30 μM, 1 h), a well-known

ERK1/2 inhibitor was used as a control. β-actin and total ERK1/2 were used as loading controls. Proteins were run on a 10 % SDS-PAGE gel. Assessment of band intensities of

(relative to total ERK1/2) by scanning densitometry. Serum-starved rat neonatal WKY12-

22 SMCs were incubated with (E) amiloride and (F) amino-caproic acid (50 and 100 μM each), inhibitors of u-PA and plasmin respectively for 30 min prior to PDGF-BB stimulation for 30 min. The data are representative of three different experiments. * p<0.05, **p0.01, *** p<0.001. Error bars represent the mean ±SE.

196

5.3.5 APMA rescues PDGF-BB-inducible Egr-1 mRNA exposed to MMP-inhibitors

APMA stimulation of VSMCs resulted in enhanced HB-EGF shedding that was inhibited by pre-incubation of BiPS (Mifune et al. 2005). To test whether APMA could restore

PDGF-BB-inducible Egr-1 mRNA in cells exposed to MMP inhibitors, rat neonatal

WKY12-22 SMCs, were preincubated with GM6001 (25 μM) or BiPS (10 μM) for 30 min, then treated with APMA (30 μM) for 1 h, followed by PDGF-BB (50 ng/ml) exposure for

30 min. APMA rescued PDGF-BB-inducible Egr-1 mRNA which had previously been inhibited by GM6001 and BiPS (Figure 5.9A-B). These data confirm that Egr-1 induction by PDGF-BB requires the intact activity of MMPs.

197

198

Figure 5.9 APMA rescues Egr-1 mRNA expression. Serum arrested rat neonatal

WKY12-22 SMCs were incubated with (A) BiPS (10 μM) for 30 min, or (B) GM6001 (25

μM) for 30 min, followed by APMA addition (30 μM) for 1 h, then stimulated with PDGF-

BB (50 ng/mL) for 30 min. SMCs were then subjected to RNA extraction and real-time

PCR analysis (Sections 4.2.3 and 4.2.4). The data are representative of three different experiments. * p<0.05, ** p<0.01, *** p<0.001. Error bars represent the mean ±SE.

199

5.3.6 PDGF-BB-inducible Egr-1 expression is mediated by EGFR

To confirm the role of PDGFRβ in PDGF-BB-inducible Egr-1, rat neonatal WKY12-22

SMCs were pretreated with AG1295, a potent inhibitor of PDGFRβ tyrosine kinase for 30 min before the addition of PDGF-BB for 30 min. As expected, AG1295 inhibited PDGF-

BB-inducible Egr-1 mRNA and protein levels (Figure 5.10A, D-E). AG1295 also blocked

ERK1/2 activation (Figure 5.10D-F). These results are in agreement with previous data demonstrating that inhibition of PDGFR-tyrosine phosphorylation by genistein, results in reduced PDGF-BB-induction of Egr-1 mRNA, which also affects messangial cell proliferation (Rupprecht12 et al. 1993). This confirms that PDGF-BB-inducible Egr-1 expression, signals through activation of the PDGFRβ.

Previous investigations have demonstrated that PDGF-BB transactivates EGFR, which in turn initiates activation of ERK1/2, an event necessary for Egr-1 up-regulation (Mineo et al. 1996; Liu and Anderson 1999; Li et al. 2000; Saito et al. 2001). It was hypothesised that PDGF-BB-inducible Egr-1 expression involves activation of the EGFR. For this purpose, we used specific inhibitors of EGFR , intrinsic kinase activity, the tyrphostin

AG1478 and PD1530355. Pre-treatment of quiescent rat neonatal SMCs with these inhibitors, followed by incubation with PDGF-BB (50 ng/mL) for 30 min, resulted in dose- dependent inhibition of Egr-1 mRNA expression as measured by real-time PCR (Figure

5.10B-C) and protein by western blot (Figure 5.10D-E). AG1478 also blocked ERK1/2 activation (Figure 5.10D-F). To provide further confirmatory evidence for EGFR in

PDGF-BB-inducible Egr-1 expression, we transfected rat neonatal WKY12-22 SMCs with

EGFR siRNA. Total EGFR protein expression was knocked down compared with cells 200

transfected with a siRNA negative control or with vehicle alone (FuGENE6 transfection agent) (Section 5.2.7) (Figure 5.10G-H). As observed in AG1478-treated SMCs, PDGF-

BB-inducible Egr-1 protein expression was reduced after silencing of EGFR, (Figure

5.10G, I) and this occurred without any reduction in PDGFRβ protein levels (Figure

5.10G, J). These results, taken together, indicate that EGFR is required for PDGF-BB- inducible Egr-1 expression.

201

202

D E F

203

H I J

204

Figure 5.10 Induction of Egr-1 by PDGF-BB is EGFR-dependent. Quiescent rat neonatal WKY 12 22 SMCs were incubated with various concentration of EGFR inhibitors

(AG1478 and PD153035) and PDGFRβ inhibitor (AG1295), for 30 min prior to PDGF-BB exposure (50 ng/mL) (30 m). (A) Real-time PCR showing Egr-1 mRNA from WKY12-22

SMCs treated with AG1295. (B) Real-time PCR showing Egr-1 mRNA from WKY12-22

SMCs treated with AG1478. (C) Real-time PCR showing Egr-1 mRNA transcripts from

WKY12-22 SMCs treated with PD153035. (D) Western Blotting analysis showing Egr-1 and ERK1/2 phosphorylation from total protein lysates of WKY12-22 SMCs treated with

AG1478 (5 μM) and AG1295 (5 μM) for 30 min followed by PDGF-BB stimulation (50 ng/mL) for 30 min. Proteins were run on a 10 % SDS-PAGE gel. Assessment of band intensities of (E) PDGF-BB-inducible Egr-1 (relative to β-actin) and (F) phosphorylation of

ERK 1/2 (relative to total ERK1/2) by scanning densitometry. (G) Western Blotting of

WKY12-22 SMCs transfected with either EGFR siRNA (0.1 μM) or siRNA negative control (0.1 μM) were stimulated with PDGF-BB (50 ng/ml) for 30 min. Proteins were run

6 % SDS-PAGE. β-actin shows unbiased loading. Assessment of band intensities for (H)

EGFR protein relative to β-actin, (I) Egr-1 (relative to β-actin) and (J) PDGFRβ levels

(relative to β-actin) by scanning densitometry. Figures are representative of at least three independent experiments. * p<0.05, ** p<0.01,*** p<0.001. Error bars represent the mean

±SE.

205

5.3.7 PDGF-BB-inducible Egr-1 expression does not require ErbB2

EGFR, is a member of the ErbB family of receptors which is composed of four closely related receptor tyrosine kinases EGFR (ErbB-1), HER2/c-neu (ErbB-2), Her 3 (ErbB-3) and Her 4 (ErbB-4) (Chan et al. 2006). To explore the role of other members of the EGFR family on PDGF-BB-inducible Egr-1 expression, rat neonatal WKY12-22 SMCs were pretreated with AG825, an ErbB-2 inhibitor, for 30 min before the addition of PDGF-BB

(50 ng/mL) for 30 min. AG825 did not downregulate PDGF-BB-induced Egr-1 mRNA and protein (Figure 5.11A-C), and did not affect the activation of ERK1/2 (Figure 5.11B,

D), suggesting that PDGF-BB-inducible Egr-1 and ERK1/2 activation are Erb-2- independent.

206

B C D

207

Figure 5.11 Role of ErbB2 on PDGF-BB-inducible Egr-1 expression. Quiescent rat neonatal WKY 12 22 SMCs were incubated with different concentrations of ErbB-2 inhibitor, AG825, for 30 min prior to PDGF-BB stimulation (50 ng/mL) for 30 min. (A)

Real-time PCR showed the effect of AG825 on Egr-1 mRNA levels. (B) Western Blotting analysis on Egr-1 and phosphorylated ERK1/2 in SMCs treated with AG825 (10 μM) prior to PDGF-BB stimulation for 30 min. Data are representative of three independent experiments. Assessment of band intensities of (C) PDGF-BB-inducible Egr-1 (relative to

β-actin) and (D) phosphorylation of ERK1/2 (relative to total ERK 1/2) by scanning densitometry. * p<0.05, ** p<0.01, *** p<0.001. Error bars represent the mean ±SE.

208

5.3.8 PDGFRβ is not required for EGFR-ligands-induction of Egr-1 expression

Previous studies (Liu et al. 2000; Kaufman and Thiel 2001), have shown that Egr-1 expression is stimulated by EGFR ligands such as HB-EGF and EGF in different cell types

(Liu et al. 2000; Kaufman and Thiel 2001). We have confirmed that these two EGFR- ligands induce Egr-1 mRNA in a time-dependent manner (Figure 5.12A-B). It was therefore investigated whether PDGFRβ is involved in EGF- or HB-EGF-induction of Egr-

1 mRNA. WKY12-22 SMCs were incubated with AG1478 (EGFR inhibitor) and AG1295

(PDGFRβ inhibitor) for 30 min before exposure to HB-EGF or EGF (10 ng/mL each) for

30 min. AG1478 inhibited EGF-and HB-EGF-inducible Egr-1 mRNA expression, whereas

AG1295 had no effect (Figure 5.12C-D), indicating that PDGFRβ is not required for EGF nor HB-EGF-inducible Egr-1 expression. The data suggest that PDGFRβ-EGFR activation is unidirectional.

209

A B

C D

210

Figure 5.12 EGFR/PDGFRβ cross talk is unidirectional. Serum-arrested rat neonatal

WKY12-22 SMCs for 24 h, were incubated with HB-EGF (10 ng/mL) (A) or EGF (10 ng/mL) (B) for 0,5, 1, 2 or 4 h. Real-time PCR graph showing Egr-1 mRNA expression over time, relative to β-actin. (C) Real-time PCR in HB-EGF (10 ng/mL) or (D), EGF (10 ng/mL)-treated SMCs for 30 min after pre-incubation with AG1295 (5 μM) or AG1478 (5

μM) for 30 min. Data are representative of three independent experiments. * p<0.05, ** p<0.01, *** p<0.001. Error bars represent the mean ±SE.

211

5.3.9 Inhibition of endogenous HB-EGF, AT II receptors or FGFR-1 do not perturb

PDGF-BB-inducible Egr-1 expression

There is growing evidence that EGFR transactivation can occur by direct activation of

EGFR tyrosine kinase activity, or by release of EGFR ligands such as HB-EGF (Prenzel et al. 1999). To determine whether endogenous HB-EGF is involved in PDGF-BB-induced

Egr-1 expression, CRM197 was utilised. CRM197 is a nontoxic mutant of diphtheria toxin that binds to, and internalises pro-HB-EGF, thus inhibiting HB-EGF cell release (Zhou et al. 2007). CRM197 did not inhibit PDGF-BB-inducible Egr-1 mRNA expression when pre-incubated for 30 min or 24 h prior to PDGF-BB exposure (50 ng/mL, for 30 min)

(Figure 5.13A-B) In this cellular model, it appears that the HB-EGF/EGFR system is not required for PDGF-BB-inducible Egr-1 expression.

Having demonstrated that in rat neonatal WKY12-22 SMCs, EGFR is required for the induction of Egr-1 by PDGF-BB, and given published evidence that G-protein-coupled receptors (GPCRs) are known to transactivate EGFR in various cell types (Saito et al.

2002; Mifune et al. 2005), as well as in SMCs (Chan et al. 2006), it was investigated whether ATII receptors (type 1 (AT1R) and 2 (AT2R) members of GPCR family) could mediate PDGF-BB-induction of Egr-1 in WKY12-22 SMCs. SMCs were treated with

Losartan (AT1R inhibitor) or PD123391(AT2R inhibitor) for 1 h prior to PDGF-BB stimulation (50 ng/mL, 30 min). In WKY12-22 SMCs treated with Losartan, levels of Egr-

1 mRNA induced by PDGF-BB remained unchanged (Figure 5.13C). WKY12-22 SMCs treated with PD123391, also failed to down-regulate this induction (Figure 5.13C). 212

Together with earlier data, this indicates that Egr-1 induction by PDGF-BB requires

PDGFRβ-EGFR, but not ATII receptors.

It has been reported that in human smooth muscle cells (HSMC), PDGF-BB induces activation of another tyrosine kinase receptor, FGFR-1, potentially by releasing FGF-2, and that FGFR-1 is required for a mitogenic response to PDGF-BB (Millette et al. 2005). To investigate the role of FGFR on PDGF-BB-induced Egr-1 expression, quiescent SMCs were incubated with different concentrations (10-20 μM ) of SU5402, a FGFR inhibitor, for

1 h, prior to stimulation with PDGF-BB (50 ng/mL) for 30 min. SU5402 did not reduce

PDGF-BB-induced Egr-1 mRNA levels (Figure 5.13D), suggesting that like endogenous

HB-EGF or ATII receptors, FGFR is not involved in the PDGF-BB-inducible Egr-1 expression.

213

A B

C D

214

Figure 5.13 PDGF-BB-inducible Egr-1 expression is ATR-FGFR-independent. Real- time PCR analysis from rat neonatal WKY12-22 SMCs pretreated with 10 μg/ml CRM197 for either (A) 30 min or (B) 24 h followed by stimulation with 50 ng/ml of PDGF-BB for

30 min. (C) Serum-starved SMCs were exposed to PDGF-BB (50ng/ml) for 30 min after pretreatment with Losartan (AT1R inhibitor, 1 μM) or PD123391 (AT2R inhibitor, 1 μM) for 1 h. (D) Quiescent SMCs were pretreated with FGFR inhibitor SU5402 (10 μM and 20

μM) for 1 h prior stimulation with PDGF-BB (50ng/ml) for 30 min. Data are representative of three independent experiments. * p<0.05, ** p<0.01, *** p<0.001. Error bars represent the mean ±SE.

215

5.3.10 PDGF-BB induces PDGFRβ-EGFR complex formation in a MMP/ADAM-

dependent manner

To determine whether PDGF-BB stimulates EGFR transactivation, quiescent rat neonatal

WKY12-22 SMCs were stimulated with PDGF-BB for various times. PDGF-BB induced

EGFR phosphorylation (pTyr1068) as early as 3 min, peaking at 5 min (Figure 5.14A-B).

These results corroborate EGFR transactivation by PDGF-BB. Previous studies reported that PDGFRβ/EGFR form heterodimers (Saito et al. 2001), however the PDGF-BB- inducible heterodimerisation is still unclear. We hypothesied that in rat neonatal WKY12-

22 SMCs, PDGFRβ and EGFR heterodimerise upon PDGF-BB stimulation. Co- immunoprecipitation analysis revealed that EGFR and PDGFRβ physically associate within

5 min of exposure to PDGF-BB (50 ng/mL) (Figure 5.14C-E).

Since this study links, for the first time, PDGF-BB-inducible Egr-1 expression with

MMPs/ADAMs, we next interrogated their role in PDGFRβ/EGFR heterodimer formation.

Rat neonatal WKY12-22 SMCs were pretreated with GM6001, BiPS or TAPI-1 for 30 min, or with APMA for 1 h, followed by PDGF-BB stimulation for 5 min. Our results showed that the MMP/ADAM inhibitors attenuated PDGF-BB-induced heterodimer formation

(Figure 5.14F, upper panel, lanes 3, 5 and 7) compared to cells pre-treated with vehicle, or the GM6001 inactive analogue (Figure 5.14F, lanes 2 and 4). Remarkably, the

MMP/ADAM activator, APMA enhanced induction of PDGFRβ/EGFR heterodimerisation

(Figure 5.14F, lane 6), revealing significant MMP/ADAM involvement in PDGF-BB driven PDGFRβ/EGFR complex formation. 216

To further corroborate that PDGF-BB-inducible PDGFRβ-EGFR activation is

MMP/ADAM-dependent, we determined the effect of GM6001, BiPS or TAPI-1 on PDGF-

BB-inducible EGFR phosphorylation at Tyr1068, a major autophosphorylation site in EGFR.

The EGFR inhibitor (AG1478), as expected, blocked EGFR phosphorylation (Figure

5.14H). PDGFRβ inhibitor (AG1295), also had similar effects (Figure 5.14H). All

MMP/ADAM inhibitors tested, blocked PDGF-BB-inducible EGFR phosphorylation

(Figure 5.14H-K), whereas the inactive counterpart of GM6001 did not. These results demonstrate that PDGF-BB induces EGFR transactivation in an MMP/ADAM-dependent manner.

217

A B

C D E

218

F G

H I

J K

219

Figure 5.14 PDGF-BB transactivates EGFR, and promotes association between

PDGFRβ/EGFR in a time- and MMP/ADAM-dependent manner. (A) Time course of

EGFR phosphorylation (p-Tyr1068) in WKY12-22 SMCs treated with PDGF-BB (50 ng/mL), with total EGFR and PDGFRβ as loading controls. (B) Assessment of band intensities. (C) Serum starved rat neonatal WKY12-22 SMCs were stimulated with PDGF-

BB (50 ng/mL) for the indicated times. Cellular extracts were immunoprecipitated with anti-PDGFRβ and subjected to Western blotting using anti-EGFR. The membranes were reprobed with anti-PDGFRβ antibody. Western Blot analyses of PDGFRβ and EGFR total protein levels were used as loading controls (last two lower panels). (D) Assessment of band intensities of PDGF-BB-inducible EGFR/PDGFRβ complex (relative to EGFR total input) and (E) PDGFRβ/PDGFRβ complex (relative to total input PDGFRβ) by scanning densitometry. (F) Quiescent SMCs were stimulated with PDGF-BB (50 ng/mL) for 5 min after pretreatment with 10 μM GM6001 together with its negative counterpart, 10 μM BiPS and 10 μM TAPI-1 for 30 min or 30 μM APMA for 1 h. Cell lysates were immunoprecipitated with anti-PDGFRβ and analysed by immunoblotting for EGFR and

PDGFR. (G) Assessment of band intensities of PDGF-BB-inducible EGFR/PDGFRβ complex (relative to EGFR total input). Western Blot showed equal loading. IP, immunoprecipitated; IB, immunoblotted. (H, I) Western Blot of EGFR Tyr1068 with total

EGFR as loading control in rat neonatal SMCs pretreated with AG1478 (5 μM), AG1295 (5

μM), GM6001 (10 μM), BiPS (10 μM) and TAPI-1 (10 μM) for 30 min followed by

PDGF-BB stimulation (50 ng/mL) for 5 min. Data are representative of three independent experiments. * p<0.05, ** p<0.01, *** p <0.001. Error bars represent the mean ±SE.

220

5.3.11 PDGF-BB-inducible WKY12-22 SMC migration is both EGFR- and MMP-

dependent

MMPs degrade extracellular matrix components, playing an important role in cell invasion in physiological and pathological processes. It has been shown that BiPS, blocks ATII- induced migration and proliferation in SMCs (Saito et al. 2002). In addition, EGFR has been linked to PDGF-BB-induced cell migration. PDGF-induced migration of fibroblasts is blocked by inhibition of EGFR phosphorylation (Li et al. 2000). Likewise, PDGFRβ inhibition by imatinib decreases PDGF-induced medulloblastoma cell migration

(Abouantoun and MacDonald 2009). To determine the effect of MMPs/ADAMs and

EGFR inhibitors on rat neonatal WKY12-22 SMCs migration and proliferation induced by

PDGF-BB, we performed an in vitro injury assay. GM6001, BiPS and TAPI-1 reduced this reparative response to injury, in cells stimulated with PDGF-BB, compared to cells treated with vehicle or serum (Figure 5.15). Neonatal WKY12-22 SMCs migration and proliferation induced by PDGF-BB was also inhibited by AG1478 (and AG1295) (Figure

5.15), indicating that MMP/ADAM and EGFR are required for PDGF-BB-dependent

WKY12-22 SMCs mitogenic and proliferative properties in vitro. We failed to demonstrate the effect of APMA on the PDGF-BB-induced SMC migration as APMA appeared to be toxic for the cells for prolonged incubation.

221

222

Figure 5.15 MMP/ADAM and EGFR inhibitors block PDGF-BB-induced SMCs migration and proliferation. Serum-arrested rat neonatal WKY12-22 SMCs were incubated with inhibitors, GM6001 (25 μM), BiPS (10 μM), TAPI-1 (10 μM), AG1478 (5

μM) and AG1295 (5 μM) or vehicle for 30 min followed by scraping and the subsequent addition of PDGF-BB (50 ng/mL). Photographs of the wound area were taken at the time of injury at 24, 48 and 72 h. Injured cells were also grown in 5 % FCS as a positive control. The shown photomicrographs were taken at 48 h. APMA was toxic at 24h (data not shown). Photographs are representative of three independent experiments.

223

5.3.12 The effect of MMP/ADAM inhibitors on PDGF-BB-inducible Egr-1 expression

is cell type-dependent

SMCs are well known for their phenotypic plasticity. The preceding data were derived using rat neonatal WKY12-22 SMCs which resemble the “synthetic” phenotype. To test whether or not PDGF-BB-inducible Egr-1 expression was MMP/ADAM and EGFR- dependent on SMCs resembling the “contractile” phenotype, we used rat aortic SMCs

(RASMCs) and human aortic SMCs (HASMCs). RASMC (Figure 5.16A) and HASMC

(Figure 5.16B) were incubated with MMP inhibitors (GM6001, BiPS and TAPI-1) for 30 min, followed by exposure to PDGF-BB for another 30 min. None of the MMP inhibitors had an effect on PDGF-BB-inducible Egr-1 mRNA, compared with rat neonatal WKY 12-

22 SMCs, despite PDGF-BB inducing Egr-1 mRNA in all three cell types. Neonatal WKY

12-22 SMCs, unlike RASMCs and HASMCs, are phenotypically similar to those found in neointimal lesions (Lemire et al. 1994), suggesting a cell subtype-dependent response.

224

225

Figure 5.16 Effect of MMP inhibitors on Egr-1 mRNA levels in other SMCs subtypes.

Quiescent (A) RASMCs and (B) HSMCs were incubated with GM6001 (GM+) (25 μM),

GM6001 inactive analogue (GM-) (25 μM), BiPS (10 μM) and TAPI-1 (20 μM) for 30 min, prior to PDGF-BB exposure for 30 min. Real-time PCR shows that MMP/ADAM inhibitors do not affect Egr-1 mRNA expression levels in either cell type. Data are representative of three independent experiments. *** p<0.001. Error bars represent the mean ±SE.

226

5.4 Discussion

This study reports that PDGF-BB induction of Egr-1 involves MMP/ADAM-dependent

ERK1/2 activation, PDGFRβ and EGFR phosphorylation. Numerous pharmacologic

MMP/ADAM inhibitors prevent Egr-1 induction, and in vitro MMP-activator potentiates

Egr-1 expression. APMA rescued Egr-1 levels after MMP inhibitor treatment. Egr-1 induction by PDGF-BB was blocked by EGFR tyrosine kinase inhibition and siRNA.

Finally, we found that EGFR and PDGFRβ associate upon PDGF-BB exposure in a

MMP/ADAM-dependent manner, and that PDGF-BB-inducible SMC repair after in vitro injury requires both EGFR and MMPs/ADAMs.

Several lines of evidence indicate rapid induction of Egr-1 expression after exposure to pathophysiological stimuli via MAP kinase signalling. This was confirmed in this study in rat neonatal WKY12-22 SMCs, in the context of PDGF-BB-inducible ERK1/2-dependent

Egr-1 expression. Besides PDGF, other growth factors, such as EGF, stimulate Egr-1 expression. EGF is one of several ligands of EGFR. EGFR ligand shedding is a pre-step required for receptor activation (Higashiyama et al. 2008), and occurs in response to stimuli, such as ATII, whereby the released ligand binds and activates EGFR, initiating intracellular signalling pathways (Higashiyama et al. 2008). A well-studied mechanism of

EGFR transactivation is the GPCR-stimulated shedding HB-EGF, known as “triple membrane-passing signalling” in which MMPs are responsible for proteolytic cleavage of pro-HB-EGF (Chan et al. 2006). The present study demonstrates a requirement for

MMP/ADAM in PDGF-BB-inducible ERK1/2 phosphorylation and Egr-1 expression.

Accordingly, this is the first report that demonstrates the requirement of MMPs/ADAMs, in 227

the induction of Egr-1. First, treatment with the pan spectrum MMP inhibitor GM600, but not its inactive analogue, GM6001-, reduced PDGF-BB-induced Egr-1 expression levels.

Second, BiPS, widely used as an MMP-2/-9 inhibitor, also blocked Egr-1 expression.

Interestingly, when SMCs were pretreated separately with specific MMP-2, MMP-3 or

MMP-9 inhibitors, Egr-1 induction by PDGF-BB was unaltered, suggesting a necessity for more than one MMP in this induction or perhaps that other MMPs are required in this induction. Third, TAPI-1, the ADAM17 inhibitor, was also shown to block Egr-1 induction. Fourth, APMA, an activator of MMPs, further enhanced PDGF-BB-inducible

Egr-1 expression and ERK1/2 activation. Finally, APMA rescued the “negative” effect of

MMP/ADAM inhibition on Egr-1 mRNA expression. The effect of MMP blockade was also evidenced as inhibition of ERK1/2 activation. Collectively, these findings indicate that

PDGF-BB-induction of Egr-1 is MMP/ADAM-dependent.

The dependency of PDGF-BB/PDGFRβ downstream signalling on the EGFR tyrosine kinase activity, but the lack of requirement of EGF/EGFR or HB-EGF/EGFR on PDGFRβ tyrosine activity needs to be clarified. Results from Habib et al. (Habib et al. 1986) have established that stimulation of COS-7 and Hs27 fibroblasts with EGF, resulted in rapid tyrosine phosphorylation of PDGFRβ. The most reasonable explanation would be that

EGFR/PDGFRβ interaction is cell type-dependent. The diverse signalling repertoire of every single transactivated receptor would be then dependent on whether other receptors are expressed in the same cell, which has implications for the downstream signalling cascade. 228

The transient and rapid increase in receptor complex formation that occurs as early as 5 minutes after exposure to PDGF-BB, correlates with the phosphorylation profile of EGFR at Tyr1068, a well-established EGFR auto phophorylation site (Downward and Parker 1984).

It is thus tempting to speculate that once EGFR is phosphorylated, it physically and rapidly interacts with the PDGFRβ. Inhibition of EGFR phosphorylation by GM6001 has been linked to GPCR-induced signalling, indicating that MMP-induced EGFR phosphorylation

(Tyr845 and 1068) is specific to GPCR signalling (Santiskulvong and Rozengurt 2003). Rat neonatal WKY12-22 SMCs treated with MMP/ADAM inhibitors (GM6001, BiPS and

TAPI-1), the PDGFRβ inhibitor (AG1295), and as expected, the EGFR inhibitor (AG1478) prevented the PDGF-BB-inducible EGFR phosphorylation at Tyr1068. Importantly, the extension in EGFR phosphorylation inhibition by AG1478 compares to that observed in inhibition of Egr-1 protein expression.

These findings further demonstrate the requirement of PDGF-BB/PDGFRβ signalling,

ERK1/2 phosphorylation and Egr-1 expression on EGFR. Interestingly, EGFR ligand- induced Egr-1 expression does not appear to be dependent upon PDGFRβ. The result indicates that EGFR does not necessarily require ligand binding for tyrosine activity and dimerisation, suggesting that EGFR signalling and EGFR transactivation need not be initiated by EGFR ligand, as our results on phosphorylation of at Tyr1068 support. This study also shows that EGFR-PDGFRβ heterodimerisation in SMCs exposed to PDGF-BB is MMP/ADAM-dependent. This result is in agreement with recent findings, in which is suggested that EGFR/ERK1/2 signalling initiated by PDGF-BB/PDGFRβ is ADAM17- dependent in non-SMCs (Mendelson et al. 2010). EGFR/PDGFRβ heterodimerisation has 229

been proposed to be ROS-dependent and Src-dependent in SMCs (Saito et al. 2001), and

PDGFRβ tyrosine kinase activity-dependent in medullobalstoma cells (Abouantoun and

MacDonald 2009).

The lack of inhibition of MMP/ADAM inhibitors on PDGF-BB-induced Egr-1 mRNA in

RASMCs as well as in HASMCs is intriguing. The response suggests a cell-type dependent mechanism but still opens a door to new research on how phenotypically different SMCs respond to environmental cues, and how the signalling cascades is also affected by this response. The physiological relevance of the PDGF-BB-MMP/ADAM-

PDGFRβ-EGFR-ERK1/2-Egr-1 axis on rat neonatal WKY12-22 SMCs, is exemplified by the blockade of PDGF-BB-inducible SMC migration and proliferation, following in vitro injury, in the presence of MMP/ADAM, EGFR and PDGFRβ inhibitors. This reparative response occurs in the damaged blood vessel wall, particularly during post-angioplasty restenosis. Egr-1 is a key vascular injury-inducible transcription factor, and the

MMP/ADAM and EGFR-dependence of its induction by key growth factors such as

PDGF-BB opens the door to new opportunities for controlling the expression and activity of this pathophysiologically relevant immediate-early gene product.

A major limitation on the use of these rat SMC lines, is that these “isolated” cell monolayers do not fully represent the complete array of signaling events of the cell type when in a complex living organism, i.e. the vessel wall not only comprises SMCs .but also

ECs, fibroblasts (Chapter 1). SMCs have been removed from their “normal environment”, 230

and there are no neighbouring cells or matrix to interact with and no blood to supply growth factors, cytokines, hormones or other important factors. Thus, to “mimic” in vivo exposure to these factors, is very challenging in in vitro settings. Likewise, rat SMC response to a certain chemical could not predict human SMC response to the drug. Many of the in vitro experiments used in this Chapter, that assess the effect of MMP/ADAM as well as EGFR inhibitors might reflect on the difficulty of correlating in vitro and in vivo effect of toxicity

(Chapter 1).

As far as the limitations of this study, in regards to specificity and efficacy of the inhibitors, in particular refers to not having demonstrated that in our cellular model, the MMPs inhibitors prevent MMP activation. In this study, this statement was assumed based on different studies (GM6001 (Poncet et al.), a pan spectrum MMP inhibitor, BiPS (Tamura et al. 1998) MMP2/9 inhibitor and TAPI-1 (Reddy et al. 2009) specific ADAM17 inhibitor.

Furthermore, AG1478 and PD15335 are known as specific EGFR kinase inhibitors, which have been shown to have no effect on the activation/phosphorylation of other kinases (Boss et al. 1997). Although we showed the inhibitory effects on phosphorylation of EGFR, we did not show the lack of effect of these pharmacologic reagents on the phosphorylation of another tyrosine. An excellent example could have been to demonstrate no effect on phosphorylation of PDGFRβ. Notwithstanding these limitations, the results obtained in this study present an opportunity to use more in vitro methodologies, which expand the underlying mechanisms of action of these inhibitors. The use of these inhibitors in our in vitro cellular model has gained new insights into signalling pathways and the mechanisms 231

implicated in Egr-1 transcriptional regulation in WKY12-22 SMCs. In vivo experiments will be imperative in validating this perception.

232

6 Conclusion and future directions: Egr-1 induction by PDGF-BB

involves MMP/ADAM-dependent ERK1/2, PDGFRβ and EGFR

phosphorylation and PDGFRβ-EGFR heterodimerisation

The work in Chapter 5 was carried out to assess the role of MMP/ADAM and EGFR on

PDGF-BB-inducible Egr-1 gene expression in SMCs. Our findings have provided the first evidence that in SMCs, Egr-1 induction by PDGF-BB requires MMPs/ADAMs and occurs via the EGFR. PDGF-BB is a major stimulant of SMC proliferation and migration in in vivo and in vitro and blockade of PDGFRβ has been shown to reduce intimal growth after arterial injury (Ostman 2004). Signalling by PDGFRs has been extensively studied and involves a myriad of transduction pathways including RAS/MEK1/2/ERK1/2, which in turn activates Egr-1 (Khachigian 2006). Egr-1 is an important transcription factor implicated in the modulation of multiple genes involved in the pathogenesis of CVD including MMPs such as MMP-1, MMP-2 and MMP-9 and EGFR.

The objective of this study was to explore the role of MMPs/ADAMs and EGFR on PDGF-

BB-triggered-expression of Egr-1 in different SMC subtypes. The data presented in this chapter demonstrates that Egr-1 induction by PDGF-BB is dependent on MMPs/ADAMs and EGFR in rat neonatal WKY12-2 SMCs (Figure 6). This conclusion chapter summarises these findings and discusses in greater depth the potential implications of these novel regulatory mechanisms in SMCs.

233

6.1 Conclusions

6.1.1 PDGF-BB-inducible ERK1/2 activation and Egr-1 expression is MMP/ADAM

dependent

When the role of MMPs/ADAMs was examined in the activation of ERK1/2 by PDGF-BB, important new findings were obtained. A broad spectrum MMP/ADAM inhibitor

(GM6001), an MMP-2/MMP-9 inhibitor (BiPS) and the ADAM-17 inhibitor (TAPI-1) were able to block ERK1/2 phosphorylation in neonatal WKY12-22 SMCs. Thus, these results are the first report to demonstrate the inhibitory effects of MMP/ADAM inhibitors on PDGF-BB-induced ERK1/2 activation. To date, a similar effect of MMP inhibitors has been found on ERK1/2 activation but only in the context of GPCR agonists such as ATII and platelet-activating factor (PAF). BiPS was shown to inhibit ATII-induced ERK1/2 in vascular SMCs . Similarly, GM6001 was shown to inhibit PAF-induced ERK1/2 activation in ovine foetal venous SMCs (Zhou et al. 2007).

Many kinds of ECMs are mainly produced by SMC in both normal arterial walls and atherosclerotic lesions. In particular, type I, III, IV, V, and VIII collagens and elastin are highly expressed in atherosclerotic lesions. The MMPs comprise a family of enzymes and play a central role in the degradation of ECM components. The expression of MMPs is related to atherosclerotic formation. The MMPs produced by SMC and macrophages are

MMP-1, 2, 3, 7, 9 and 12 in the arterial wall, and have their highest expression in atherosclerotic lesions. Although MMP involvement in pathology is more than simple excessive matrix degradation, or an imbalance between them and their specific tissular inhibitors (TIMPs), MMP inhibition may be of therapeutic benefit, so synthetic MMPs 234

inhibitors had been developed and are currently under clinical testing. In atherosclerotic plaques, MMP-3 was detected in both smooth muscle cells and macrophages. The production of MMP-1, MMP-2, and MMP-9 are also seen primarily in macrophages and smooth muscle cells in the lesion. They are thought to participate in weakening the connective tissue matrix in the intima, which leads to plaque rupture, acute thrombosis, and smooth muscle cell proliferation and migration.

It is well known in the literature that ERK1/2 activation is essential for Egr-1 expression

(Khachigian et al. 1996; McCafrrey et al. 2000; Wang et al. 2006; Wang et al. 2009). It was further demonstrated in this thesis, that the same MMP/ADAM inhibitors that decreased PDGF-BB-induced ERK1/2 activation could also inhibit PDGF-BB-induced

Egr-1 expression. In agreement with the MMP/ADAM inhibitor studies, activation of

MMPs/ADAMs by APMA (a specific pharmacologic activator, shown to enhance HB-EGF shedding (Mifune et al. 2005)), not only enhanced Egr-1 induction by PDGF-BB but also rescued Egr-1 mRNA expression in PDGF-BB-stimulated WKY12-22 SMCs previously exposed to the MMP-2/MMP-9 inhibitor (BiPS) or the pan-MMP/ADAM inhibitor

(GM6001). The involvement of MMP/ADAM activity in Egr-1 expression is a significant finding as current literature examining the effects of MMP/ADAM inhibitors on Egr-1 expression is limited. Previously, one study examined the links between MMP activity and

Egr-1 expression in response to induction with the peptide hormone, arginine vasopressin

(AVP) (Fuentes et al. 2008). It was demonstrated that in A10 cells, a subtype of SMC,

AVP induction of Egr-1 occurs via RAS/MEK/ERK1/2-dependent signalling pathway. 235

However, in this case, treatment with the pan-MMP inhibitor (GM6001) did not downregulate Egr-1 expression. The lack of involvement demonstrated in this case may arise from differences in the signalling pathways used by stimuli such as growth factors

(PDGF-BB), GPCR agonists (ATII, PAF) and peptide hormones (AVP), or may originate from inherent differences between cell types or even cell subtypes.

Interestingly in this thesis, despite the inhibitory effect of the MMP-2/MMP-9 inhibitor

(BiPS) on PDGF-BB-induced Egr-1 expression, specific inhibitors targeting MMP-2,

MMP-3 or MMP-9 individually had no effect. This effect is puzzling but builds on other study indicating that BiPS could inhibit ATII-induced EGFR transactivation whereas individuals MMP-2, -3 and -9 inhibitors fail to elicit any effect (Saito et al. 2002). The authors hypothesised that BiPS could also act as an ADAM inhibitor, however this hypothesis was not tested (Saito et al. 2002). BIPS is regarded as a highly selective inhibitor of type IV collagenases or gelatinases. BiPS has been shown to have no effect on other MMPs such as MMP-1, -3, -7, angiotensin-converting enzyme (ACE), neutral endopeptidase (NEP) and endothelin-converting enzyme (ECE) (Tamura et al. 1998). The basement membrane is composed mainly of type IV collagen. The disruption of vascular basement membranes by type IV collagenase (MMP-9 and MMP-2) may play an important role in triggering the process of tumor growth. The results in this thesis suggest that type

IV collagenases maybe required for Egr-1 induction by PDGF-BB as demonstrated by the lack of effect of the MMP-3 inhibitor on this induction. MMP-3 is also known as stromelysin, which degrades non-collagen matrix proteins such as fibronectin, laminin and aggregan. However, it is intriguing that individual inhibitors to MMP-2 and MMP-9 did 236

not affect Egr-1 expression. This might be due to the fact that both MMP-2 and MMP-9 posses a similar catalytic effect on collagen type IV, and within this cellular model both

MMPs need to be simultaneously inhibited in order to block PDGF-BB-inducible Egr-1 expression. In addition, more research on the specific MMPs involved in Egr-1 expression is required (refer to 6.2.2). MMPs/ADAMs are thus required for PDGF-BB activation of

ERK1/2/Egr-1 signalling, however further study is required to more specifically define the proteases involved.

6.1.2 PDGF-BB-inducible Egr-1 expression in neonatal WKY12-22 requires EGFR

EGFR can be transactivated by PDGF-BB stimulation and that EGFR transactivation is required for PDGF-BB-stimulated fibroblast cell migration (Li et al. 2000). In addition, other investigations have demonstrated a positive regulatory role for EGFR in AVP induction of Egr-1 in A-10 SMCs (Fuentes et al. 2008), and H. pylori-activation of Egr-1 in

AGS gastric epithelial cells (Keates et al. 2005). This thesis has now demonstrated that blockade of EGFR tyrosine activity via both pharmacological and molecular approaches

(AG1478, PD153035 and siRNA) significantly inhibits phosphorylation of ERK1/2 and the downstream expression of Egr-1 in rat neonatal WKY12-22 SMCs and confirmed that

PDGF-BB signals through PDGFRβ. Thus, it appears that signalling activation initiated by

PDGF-BB/PDGFRβ, requires EGFR to activate downstream signalling pathways such as

ERK1/2, leading to expression levels of Egr-1.

A number of ligands able to transactivate EGFR have been identified, of which HB-EGF and EGF have been extensively studied and shown to induce Egr-1 expression (Miyagawa 237

et al. 1995; Kaufman and Thiel 2001). The dependency of PDGF-BB/PDGFRβ on EGFR led us to investigate whether this functional interaction was bi-directional, that is, whether

PDGFRβ blockade could also influence EGF/EGFR signalling. However, no inhibition of

HB-EGF or EGF-induced Egr-1 expression was observed. Importantly, findings by Liu et al. (Liu and Anderson 1999) demonstrate that PDGFRβ was not phosphorylated by EGF, despite PDGF-BB being able to phosphorylate EGFR. Together these data suggest that (i)

PDGFRβ/EGFR transctivation is unidirectional and (ii) a plethora of downstream molecules might be differentially activated by the growth factors PDGF-BB and EGF.

As previously mentioned, GPCR agonists are known to activate MMPs/ADAMs. Once activated, they cleave EGFR ligands, releasing their soluble and mature form, which is now ready to bind the EGFR (Prenzel et al. 1999). In order to more specifically describe the

EGFR transactivation by PDGF-BB, it was investigated whether PDGF-BB could induce release of EGFR ligands, specifically HB-EGF. However, treatment with CRM-197 (an antagonist of HB-EGF) prior to PDGF-BB-stimulation did not affect Egr-1 mRNA expression levels , suggesting that other EGFR ligand(s) may act as mediators of EGFR transactivation. However, in order to determine whether PDGF-BB induction of ERK1/2 phosphorylation and Egr-1 expression is due to PDGF-BB-induced endogenous HB-EGF release, more comprehensive studies are required (Section 6.2.5).

As EGFR is activated by multiple stimuli, it might be assumed that PDGF-BB transactivation of EGFR could have a distinctive character compared to EGFR transactivation by GPCR, whereby shedding of EGFR ligands (HB-EGF and EGF) by 238

ADAMs has been confirmed (Prenzel et al. 1999; Pierce et al. 2001; Yan et al. 2002).

Transactivation of EGFR has been studied in the context of multiple stimuli. Interestingly, when GPCR agonists are used, EGFR phosphorylation occurs at early time points and this rapid induction is comparable to that initiated by EGFR ligands. Daub et al. (Daub et al.

1997) have described the phosphorylation pattern of EGFR in the context of GPCR agonists and EGFR ligands. EGFR phosphorylation in the presence of GPCR agonists such as Thrombin and Lysophosphatidic acid (LPA), was shown to occur as early as 2 min, peaking at 5 min, in various cell types. In contrast, PDGF-BB transactivation of EGFR has been shown to occur as early as 3 min, with maximum phosphorylation levels occurring at

10 min in fibroblasts (Li et al. 2000), and also in rat aortic SMCs (Saito et al. 2001).

Intriguingly, in this study EGFR transactivation by PDGF-BB in rat neonatal WKY12-22

SMCs peaked at 5 min, which more closely resembles the rapid EGFR transactivation characteristic of GPCR agonists. As described earlier, the rat neonatal WKY12-22 SMCs resemble the “synthetic” phenotype that are more responsive to growth factors compared to the “contractile” phenotype SMCs (Majesky et al. 1992). Interestingly the SMCs used by

Saito et al. (Saito et al. 2001) reflect the later phenotype. Thus it may be that SMC subtype and whether “contractile” or “synthetic” may influence the mechanisms of PDGF-BB- inducible EGFR transactivation. This needs to be further investigated but may provide important information on how proliferative SMC phenotypes in atherosclerosis or restenosis may be targeted.

239

6.1.3 PDGF-BB promotes cross-talk between PDGFRβ and EGFR in neonatal

WKY12-22 SMCs

PDGFRβ and EGFR have been shown to physically associate (Saito et al. 2001;

Abouantoun and MacDonald 2009). However, whether there is increased association between PDGFRβ/EGFR following PDGF-BB exposure is unclear. Saito et al. (Saito et al.

2001) showed that PDGFRβ/EGFR heterodimers exist independently of PDGF-BB stimulation in rat aortic SMCs, whereas Abouantoun et al. (Abouantoun and MacDonald

2009) showed that PDGFRβ/EGFR heterodimers formed only after PDGF-BB stimulation in medulloblastoma cells, which over-express EGFR (Abouantoun and MacDonald 2009).

Results presented in this thesis reflect the latter data with PDGFRβ/EGFR heterodimer association occurring upon PDGF-BB stimulation. Again, whether these differences are a result of cell type specificity or cell type differences in EGFR expression in SMCs remain to be investigated.

The role of signalling pathways and molecules involved in PDGFRβ/EGFR heterodimer formation is emerging. ROS and Src are reported to be important in the stabilisation of the

PDGFRβ/EGFR complex in SMCs (Saito et al. 2001). Interestingly, when the role of tyrosine kinase activity of the PDGFRβ was studied, it was found that AG1295 (a compound which inhibits PDGFRβ autophosphorylation) had no effect on PDGF-BB- induced PDGFRβ/EGFR complex formation nor the transactivation of EGFR in rat aortic

SMCs (Saito et al. 2001). Intriguingly however, two other inhibitors of PDGFRβ tyrosine kinase activity, Imatinib and Sunitib, disrupted PDGF-BB-inducible PDGFRβ/EGFR 240

complex formation as well as EGFR transactivation in medulloblastoma cells (Abouantoun and MacDonald 2009; Abouantoun et al. 2010).

Additionally, in this thesis, we have reported for the first time, that MMPs/ADAMs play a role in PDGF-BB-induced PDGFRβ/EGFR complex formation in rat neonatal WKY12-22

SMCs. However, more research is required in order to understand how MMP/ADAM activity helps dimer formation and for instance whether EGFR ligand (shedding of ligands by MMPs/ADAMs) is required for dimer formation.

6.1.4 Inhibitors to MMPs/ADAMs and EGFR attenuate PDGF-BB-induced

migration and proliferation in WKY12-22 SMCs

MMP inhibitors have been shown to prevent SMC migration in the presence of IL-1β

(Kong et al. 2010) and ATII (Yang et al. 2005) and recently, it has been reported that

Marimastat, a broad spectrum MMP inhibitor, could decrease PDGF-BB-induced migration of fibroblasts (Mendelson et al. 2010). Similarly, in wounding assays performed in this thesis have now demonstrated that PDGF-BB-inducible WKY12-22 SMC migration and proliferation is blocked by MMP/ADAM inhibitors. Further to this, inhibition of EGFR also blocked PDGF-BB-induced SMC migration and proliferation. These results support a significant physiological role for the MMP/ADAM-dependent PDGF-BB signaling through

PDGFRβ/EGFR complex formation and ERK1/2-Egr-1 pathway activation in rat neonatal

WKY12-22 SMCs. 241

6.1.5 Egr-1 induction by PDGF-BB is MMP/ADAM and SMC type-dependent

This study has demonstrated that the different SMCs share similarities (Egr-1 up-regulation by PDGF-BB), but also have important differences in the response to MMP/ADAM inhibitors upon PDGF-BB stimulation. Inhibitory effects of the pan-MMP inhibitor

(GM6001), the MMP-2/MMP-9 (BiPS) and the ADAM-17 inhibitor (TAPI-1) on PDGF-

BB-inducible Egr-1 mRNA expression, is shown to be specific to SMCs derived from the aorta of rat neonatal (WKY12-22) but not to SMCs derived from the aorta of rat adult

SMCs (RASMCs) or SMCs derived from the media of normal human, fibrous plaque-free aorta (HASMCs). Those rat neonatal WKY12-22 SMCs represent the “synthetic” phenotype, whereas RASMCs and HASMCs resemble the “contractile” phenotype. As previously described (Chapter 1, Section 4.1.3 and 4.2.5), SMCs have the ability to modulate their phenotype during atherogenesis, when they migrate from the media to the intima, where they proliferate and accumulate in the arterial wall (Majesky et al. 1990). It is now hypothesised that modifications in phenotype between SMCs, may differentially contribute to arterial remodelling during development, repair and disease (Majesky et al.

1990). Thus, these results are in line with the observations previously described in this thesis, whereby, the response to ATII in the context of PDGF-C induction in SMCs is phenotype-dependent (Chapter 3). Although Egr-1 is expressed by numerous stimuli, it could now be hypothesized that its transcriptional control is also differently controlled in

SMCs from neonatal versus adult origin. As far as the Egr-1-dependence on MMPs, it is important to mention, that the divergent response to MMP inhibitors (i.e. the observed lack of effect of MMP inhibitors in the PDGF-BB-inducible Egr-1 mRNA expression in rat aortic and human aortic SMCs compared to its surprising inhibition in rat neonatal SMCs) 242

might also be explained by the fact that contractile and synthetic SMCs differentially express genes. Particularly, synthetic SMCs (neonatal origin) show up-regulation of genes related to ECM deposition, which might imply up-regulation of MMP in order to degrade the basement membrane around SMCs. Thus, this distinction might clarify the lack of inhibition of Egr-1 by MMP inhibitors in contractile SMCs. However, this theory needs further validation comparing MMP expression and activity in the two phenotypes (refer to

Chapter 6; 6.2.3 and 6.2.6).

243

Figure 6 Egr-1 induction by PDGF-BB involves MMP/ADAM-dependent ERK1/2,

PDGFRβ, EGFR phosphorylation and PDGFRβ/EGFR heterodimerisation. Model representing the findings presented in Chapter 4: PDGF-BB binds to PDGFRβ, which in turn transactivates the EGFR to promote heterodimer formation. This PDGFRβ/EGFR axis signals via ERK1/2 to activate Egr-1 gene expression. Egr-1 is critically dependent on

EGFR and MMPs/ADAMs activity, but not on an ErbB2. Inhibition of PDGF-

BB/PDGFRβ/EGFR/ERK1/2/Egr-1 axis is required for neonatal WKY12-22 SMCs proliferation and migration. Dashed lines represent ongoing work investigating

MMP/ADAM activation by PDGF-BB and release of currently unknown EGFR ligands.

Modified from Saito et al. (Saito et al. 2001). 244

6.2 Future directions

This study has provided new knowledge on the signalling events initiated by PDGF-

BB/PDGFRβ in SMCs. In this thesis, we have demonstrated that activation of ERK1/2 and

Egr-1 expression by PDGF-BB is regulated by MMP/ADAM activity and the EGFR in neonatal WKY12-22 SMCs. Moreover, PDGF-BB is shown to up-regulate

PDGFRβ/EGFR complex formation in a time and MMP/ADAM-dependent manner and shown a downstream physiological effect on cell activity. Notwithstanding the contribution that these results have provided into these signalling events between PDGF-

BB and Egr-1, this work has raised several new questions which are discussed below.

6.2.1 Is PDGF-BB-induced c-fos expression MMP/ADAM-EGFR-dependent?

This study demonstrated the requirement of MMP/ADAM and EGFR for PDGF-BB activation of ERK1/2 and induction of Egr-1. Egr-1 belongs to the early immediate gene

(IEG) family which also comprises the proto-oncogene c-fos (Shunichi and Tsumiyama

2009). Like Egr-1, c-fos expression is mediated through serum response elements (Yang et al.) within its promoter and serum response factors (Durchdewald et al. 2009). These SRE are targeted by multiple signalling pathways, particularly MAPK (Cahill et al. 1996).

PDGF-BB has been shown to activate a c-fos-dependent promoter via ERK1/2 activation

(Li et al. 2010). Thus, exploring the role of MMPs/ADAMs and EGFR on PDGF-BB- inducible c-fos expression will provide valuable new information. It would be particularly interesting to determine whether c-fos is activated via the same signalling cascade (PDGF-

BB/PDGFRβ/EGFR/ERK1/2) as Egr-1, as part of perhaps a novel signalling pathway by which IFG expression is controlled. 245

6.2.2 Which MMP(s)/ADAM(s) are controlling PDGF-BB-inducible Egr-1

expression?

Although our data solidly presents the dependency of Egr-1 on MMP/ADAM activity, it will be essential to name the MMP(s)/ADAM(s) implicated in this process, as it may represent a potential new tool in the treatment of atherosclerosis. Knock-out studies utilising cells deficient in a number of MMPs/ADAMs will be required to find the

“missing” piece of this intriguing puzzle in an inhibitor-free cellular system. Possible candidates would be MMP-2 , MMP-9 and ADAM-17 knock-out cells. In these studies many pharmacologic MMP/ADAM inhibitors were used, however, it cannot be excluded that there are potential non-specific or indirect actions of these inhibitors on gene expression. Thus, a more direct approach would include use of siRNA and tissue inhibitor of metalloproteinase (TIMP)-expressing plasmids which will further validate our conclusion on the role of MMPs/ADAMs on the PDGFRβ/EGFR-ERK1/2-Egr-1 signalling cascade initiated by PDGF-BB in neonatal WKY12-22 SMCs.

6.2.3 Do different SMCs have dissimilar expression pattern of MMPs, ADAMs,

EGFR and PDGFRβ?

This chapter demonstrates that MMP/ADAM-inducible Egr-1 down-regulation is SMC subtype dependent. As discussed in Chapter 1 (also Section 5.1.3), SMCs represent a heterogeneous population and therefore respond differently to stimuli (Owens et al. 2004).

It is intriguing that MMP/ADAM inhibitors did not affect PDGF-BB-induced Egr-1 mRNA expression in rat aortic and human SMCs, but was a phenomenon only observed in rat neonatal WKY12-22 SMCs. Thus, studies comparing basal expression levels of MMPs 246

and ADAMs as well as EGFR and PDGFRβ between these SMCs would be beneficial as it will provide a better understanding of expression patterns between contractile and synthetic phenotypes which in turn will reflect on the unresponsiveness to different stimuli.

In this study, it is shown that inhibition of PDGFRβ tyrosine kinase activity with the

AG1295 compound, results in blockade of ERK1/2 activation and Egr-1 gene expression, suggesting that PDGF-BB-induced ERK1/2 phosporylation and Egr-1 induction is

PDGFRβ-dependent. It is, however, of extreme value to validate the concept of PDGFRβ phosphorylation by PDGF-BB in our cellular model. This is important given the involvement of MMP/ADAM and EGFR in the PDGF-BB/PDGFRβ downstream signalling pathway.

6.2.4 Can PDGFRβ heterodimerise with other EGFR isoforms ?

Activation of EGFR by GPCRs leads to formation of EGFR (ErbB1) monomers. However

EGFR can also form heterodimers with other members of the EGFR family, particularly with ErbB4 (Sanderson et al. 2006). Further studies employing siRNA to ErbB4 would be beneficial to give: (i) more information on PDGFRβ/EGFR heterodimerisation between isoforms, since it will provide information on whether ErbB4 is also important in PDGF-

BB-induced PDGFRβ/EGFR heterodimerisation, and (ii) more insight into how signal transduction initiated by PDGFRβ may influence other members of the EGFR family.

247

6.2.5 Does PDGF-BB stimulation induce EGFR ligand (s) release?

This work demonstrated that PDGF-BB transactivates the EGFR, and that stimulates heterodimer formation between PDGFRβ and the EGFR. It could therefore be speculated that the transactivation of the EGFR by PDGF-BB may be a consequence of PDGF-BB- induced MMP/ADAM activation. This in turn would stimulate the release of EGFR- ligands, ultimately being responsible for the cross-talk between PDGFRβ/EGFR and its effect on downstream signalling activation. Having now demonstrated that CRM197, a molecule which binds to HB-EGF and prevents its activation of EGFR, did not inhibit Egr-

1 mRNA levels in WKY12-22 SMCs exposed to PDGF-BB, it appears that PDGF-BB does not stimulate HB-EGF, however cautious interpretation of this result needs to be made.

Various approaches could potentially help to clarify these findings. For example, by (1) measuring expression levels of EGF, HB-EGF, TNF-α and TGF-α in rat neonatal WKY12-

22 SMCs; (2) using expression constructs for alkaline phosphatase (AP)-tagged-EGFR ligands including EGF, HB-EGF, TNF-α and TGF-α would provide information on EGFR- ligands shedding upon PDGF-BB stimulation; (3) utilising ELISA experiments to determine whether PDGF-BB stimulates the release of EGFR ligands (TGF-α, EGF, HB-

EGF, AR) into the culture medium; and (4) using neutralising antibodies to EGFR ligands could determine whether ERK1/2 activation and Egr-1 expression is due to PDGF-BB- induced endogenous EGFR ligand release.

6.2.6 Does PDGF-BB activate MMP activity?

This study has shown that inhibitors (GM6001, BiPS and TAPI-1) and a stimulant (APMA) of MMP/ADAM activity negatively and positively affect PDGF-BB-induced 248

PDGFRβ/EGFR complex formation, as well as ERK1/2 activation and Egr-1 expression.

However, in order to completely validate the concept of dependency of PDGF-

BB/PDGFRβ downstream signals on MMP/ADAM activity, the ability of PDGF-BB to stimulate MMP/ADAM activity in this cell type needs to be explored. Gelatin zymography is an electrophoretic technique based on SDS-PAGE which is used for the detection of gelatin-degrading proteases, including MMP-2 and -9 (Newby 2005; Newby 2006). On the other hand, activation of ADAM-17 will be detected using antibodies specifically recognizing the phosphorylated form of ADAM-17 (Edwards et al. 2008). Another approach will be the use of AP-tagged substrates (TGFα-AP, TNFα-AP, HB-EGF-AP) for

ADAM-17, and stimulate WYY12-22 SMCs with PDGF-BB (50 ng/mL) for various times.

These experiments will expand the current knowledge of PDGF-BB-induced MMP/ADAM activity in SMCs.

6.2.7 Do MMP/ADAM inhibitors block EGFR phopshorylation at Tyr845?

Studies on the phosphorylation status of another important tyrosine residue in the EGFR,

Tyr845, have demonstrated that this tyrosine is a target of Src (Bishayee et al. 1999).

Importantly, Src is known to “stabilise” PDGFRβ/EGFR complex formation upon PDGF-

BB stimulation (Saito et al. 2001). The PDGFRβ/EGFR heterodimer is disrupted in rat aortic SMCs exposed to PP2 (a Src kinase inhibitor) suggesting that Src is required to maintain heterodimer formation (Saito et al. 2001). Thus, studies on the phosphorylation profile of EGFR, Tyr845, will be of extreme value as it might give us more detailed information on how PDGF-BB transactivates the EGFR, and importantly will add new information on the PDGFRβ/EGFR complex pathway. 249

6.2.8 Can Förster Resonance Energy Transfer (FRET) experiments demonstrate the

physical interaction between PDGFRβ/EGFR upon PDGF-BB stimulation and

confirm a role of MMP/ADAM in this complex formation?

It has been established that EGFR (Mineo et al. 1996) and PDGFR (Liu et al. 1996) are concentrated in caveolae or lipid rafts at the cell surface. Furthermore, localisation of these two receptors to this narrow cell membrane structure may facilitate their interaction

(Milenkovic et al. 2003). Lipid rafts are dispersed by cholesterol depletion (Mineo et al.

1996). Depletion of cholesterol in Müller cells decreases the mitogenic effect of PDGF suggesting the dependence of the PDGFR on EGFR (Milenkovic et al. 2003). This result supports the assumption that the two receptors are expressed in close spatial vicinity in the

Müller cell membrane.

FRET is a technique used to investigate a variety of biological phenomena that produce changes in molecular proximity. Measuring FRET enables the intracellular locations of the molecular interactions to be determined. Results presented in this thesis show that the

PDGF-BB-induced PDGFRβ/EGFR complex is disrupted by MMP/ADAM inhibitors.

Thus, FRET analysis using labelled-protein constructs, e.g. cyan fluorescent protein (CFP)-

EGFR and yellow fluorescent protein (YFP)-PDGFRβ, will provide us with more detailed information including time-dependent complex dissociation or localisation of

PDGFRβ/EGFR within the plasma membrane. Therefore determining whether PDGFRβ and EGFR are co-localised in the cell membrane may help shed light on the mechanisms behind their “close” interaction, particularly when PDGFRβ signalling is dependent on

EGFR. 250

6.2.9 Do second messengers play a role in PDGF-BB-induced EGFR

transactivation?

EGFR transactivation by GPCR appears to require transduction pathways generated by second messengers, including elevation of intracellular Ca2+, protein kinase C and generation of ROS (Mifune et al. 2005). PKC appears to play an important role in transactivation of GPCRs (Milenkovic et al. 2003). In addition, the crucial role of Ca2+ and

ROS in GPCR-inducible EGFR transactivation has been elucidated (Mifune et al. 2005).

Further study on this aspect is necessary for understanding the complex functionality of the

EGFR system in the context of PDGF-BB/ PDGFRβ signalling.

6.2.10 Could Egr-1 overexpression overcome the suppressive effect of MMP/ADAM

inhibitors on SMC migration and proliferation upon PDGF-BB stimulation?

In this work it is demonstrated that MMP/ADAM inhibitors block PDGF-BB inducible

Egr-1 expression as well as proliferation and migration of rat neonatal WKY12-22 SMCs.

Previous studies have demonstrated that antisense and siRNA to Egr-1 can inhibit serum- inducible regrowth of rat neonatal WKY12-22, and human aortic SMCs respectively, in an in vitro scrape injury model (Fahmy and Khachigian 2002; Fahmy and Khachigian 2006).

It may therefore be assumed that the negative effect of MMP/ADAM inhibitors on migration and proliferation observed in this thesis could be a consequence of the downregulation of Egr-1 expression. Thus, it is hypothesised that overexpression of Egr-1 may rescue the anti-proliferative and anti-mitogenic effects of the MMP/ADAM inhibitors on SMCs. To verify this statement, Egr-1 overexpression vectors (Adeno-Egr-1) could be 251

utilised. This study would give us more insights regarding the role of Egr-1 on SMC proliferation and migration.

6.2.11 Can EGFR ligands induce PDGFRβ/EGFR assembly?

Results in this thesis have demonstrated that PDGF-BB promotes PDGFRβ/EGFR heterodimer formation. It would be interesting to establish whether the addition of recombinant EGFR ligands including HB-EGF, EGF, TGF-α, AR, stimulates

PDGFRβ/EGFR assembly as well as ERK1/2 activation and Egr-1 expression. These experiments would help us to clarify whether PDGFRβ/EGFR assembly is bi or unidirectional mechanism.

6.2.12. Is ADAM/MMP-dependent Egr-1 expression functional in other cell types, or

is it confined to “synthetic” SMCs?

Matrix metalloproteinases (MMPs) have important roles in several cancer-supporting cellular processes, such as extracellular matrix (ECM) remodeling, angiogenesis, apoptosis, epithelial-to-mesenchymal transition and cell proliferation. To test the validity of the findings of this thesis, key experiments (looking at the ADAM/MMP-dependence of

PDGF-BB-inducible Egr-1 expression using inhibitors) would be performed in non SMCs such as EC, fibroblasts and cancer cells (MDA-MB-231, MDA-468 and MCF-7 breast cancer cells, U-87MG, U-343 and U-118-gliobastoma cells ).

252

6.2.13 To demonstrate that Egr-1 induction is both ADAM/MMP in animal models

Rat ballon catheter experiments, a well-established animal model of vascular SMC injury, could be used to validate the findings by which in PDGF-BB-treated neointimal SMCs

(which are represented by pups WKY12-22 SMCs used in in vitro experiments, Egr-1 expression is blocked by MMP inhibitors. Balloon injury to rat carotid arteries results in instant denudation of the endothelium, SMC hyperplasia is observed within 4 days and neoitima formation occurs 14 days after. The experiment will examine the role of Egr-1 expression with or without the infusion of PDGF-BB together with or without the delivery of MMP inhibitors, at doses previously used in rodents (eg. 100μg/kg of GM6001, 5mg/kg of TAPI-1, intra-peritoneal). Injured arteries will be collected over a time course (i.e. 2, 8,

12, 24h and then daily for 14 days), homogenized and used in PCR experiments to determine the optimal time for the determination of Egr-1in the tissue. These experiments will be the performed with inhibitors that disturb this cascade at the optimal time. In another group of animals, carotid arteries will be perfusion fixed with formalin and used for

IHC analysis for Egr-1 expression. This will prove the spatial localization of Egr-1 in

SMCs of injured vessels and not in SMCs of intact vessels (which represent the adult phenotype used in the in vitro experiments) and that the inhibitors block this pathway as they do in vitro”.

253

7 Final conclusion of this thesis

In this thesis, the molecular mechanisms mediating expression of genes potentially implicated in atherosclerosis were examined. First, the novel mechanism by which ATII activates transcription of PDGF-C. This thesis has illustrated that ATII, via AT1R, up- regulates PDGF-C gene expression and that induction of this gene is regulated by the transcription factor Egr-1, a mechanism observed in rat neonatal WKY12-22 but not in rat adult WKY2-22M SMCs. This further exemplifies differences observed between SMC subtypes. At the present time, there is very limited information about the factors controlling PDGF-C expression at the level of the promoter or its involvement in atherosclerosis. However, the gratifying aspect of these findings is that they provide the first mechanism demonstrating that ATII induces PDGF-C transcriptional activation, which adds to the growing body of knowledge on the role of PDGF family members in disease.

Secondly, the mechanisms controlling PDGF-BB induction of Egr-1 expression were studied. In particular, the functional role of MMPs/ADAMs and EGFR were investigated.

These studies indicate that the PDGF-BB/PDGFRβ axis is dependent on MMPs/ADAMs and EGFR to initiate downstream signalling cascades, specifically ERK1/2 activation and

Egr-1 gene expression. Mechanistically, it has been shown that PDGF-BB stimulates

PDGFRβ/EGFR association in an MMP/ADAM dependent manner, demonstrating that

MMPs/ADAMs are critical factors regulating receptor heterodimerisation, indicating novel and important insights into PDGF-BB/PDGFRβ signaling cascade. Furthermore, this thesis 254

has demonstrated that MMP/ADAM/EGFR involvement also regulates rat neonatal

WKY12-22 SMC proliferation and migration. Thus, this axis of PDGF-

BB/PDGFRβ/MMP/ADAM/EGFR may be a common theme in mitotic signal transduction.

Results presented in this thesis, demonstrate ADAM/MMP-dependence of PDGF-BB- inducible Egr-1. This result would suggest that MMPi could be use as potential treatment for atherosclerosis. To date, only three MMPi’s have been assessed clinically for cardiovascular indications: doxycycline (MIDAS – Metalloproteinase Inhibition with submicrobial doses of Doxycycline to prevent Acute coronary Syndromes and doxycycline

Hyclate-PeriostatR for acute coronary syndrome), batimastat (BRILLIANT–EU –

Batimastat antiRestenosis trial utiLiz-Ing the BiodivYsio locAl drug delivery PC stent or broad-spectrum inhibitor batimastat), eluting stents for restenosis, and PG-116800

(PREMIER – PREvention of MI Early Remodeling) for the prevention of postischemic left ventricular dilation). In all three studies, there was no significant difference between the drug and non-drug treated groups. Moreover, MMP inhibitors, which are potent zinc chelators such as batimastat and PG-116800, appear to induce the musculoskeletal syndrome, manifested as a tendonitis-like fibromylagia, the major side-effect limiting dose selection (Amălinei et al. 2010). These investigations together with the finding presented in this thesis deserve and encourage more tireless research to be able to overcome the side- effect of MMPi in order to utilise them as treatment for atherosclerosis.

This thesis therefore has generated important findings on novel signaling mechanisms by which the expression pattern of genes involved in atherosclerosis are affected. General observations to be drawn from this thesis are that PDGF and Egr-1 could be potential 255

targets for therapeutic approaches in the treatment of atherosclerosis. Although this is an immune-inflammatory and complex molecular process, the results presented here continue to expand knowledge on this pathological condition and provide direction for more research to further understand mechanisms of SMC proliferation and migration and potentially to develop better diagnostics with the ultimate goal of preventing atherosclerosis and other vascular occlusive diseases.

256

8 References Abe, J., J. Deguchi, Y. Takuwa, K. Hara, Y. Ikari and T. Tamura (1998). "Tyrosine phosphorylation of platelet derived growth factor beta receptors in coronary artery lessions: implications for vascular remodelling after directional coronary atheroctomy and unstable angina pectoris " Heart 79(4): 400-406. Abouantoun, T., R. Castellino and T. MacDonald (2010). "Sunitib induces PTEN expression and inhibits PDGFR signaling and migration of meduloblastoma cells." J Neurooncol 15: 254-258. Abouantoun, T. and T. MacDonald (2009). "Imatinib blocks the migration and invasion of medulloblastoma cells by concurrently inhibiting activation of platelet-derived growth factor receptor and transactivation of epidermal growth factor receptor." Mole Canc Ther 8(8): 1137-1147. Adams, J. (2002). "Molecular organisation of cell-matrix contacts: essential multiprotein assemblies in cell and tissue function." Expert Rev Mol Med 4: 1-24. AIH&W (2005). "Cardiovascular Health." Australian Institute of Health and Welfare 2005. Akishita, M. (1999). "Expression of the AT2 receptor developmentally programs extracellular signal-regulated kinase activity and influences fetal vascular growth." J Clin Inves 103: 63-71. Almholt, K., A. Kuncker-Jensen and O. Laerum (2008). "Metastasis is strongly reduced by the matrix metalloproteinase inhibitor Galardin in the MMTV-PymT transgenic breast cancer model." Mol Cancer Ther 7(9): 2758-2767. Amălinei, C., I. Căruntu, S. Giuşcă and R. Bălan (2010). "Matrix metalloproteinases involvement in pathologic conditions." Rom J Morphol Embryol 51(2): 215-228. Amano, S., S. Yamagishi, Y. Inagaki and T. Okamoto (2003). "Angiotensin II stimulates platelet-derived growth factor-B gene expression in cultured retinal pericytes through intracellular reactive oxygen species generation." Int J Tissue React 25(2): 51-55. Amour, A., C. Knight and A. Webster (2002). "The enzymatic activity of ADAM-8 and -9 is not regulated by TIMPs." FEBS Lett 524(1-3): 154-158. Andrae, J., R. Gallini and C. Betsholtz (2008). "Role of platelet-derived growth factors in physiology and medicine." Genes Dev 22: 1276-1312. Anliker, B. and J. Chun (2004). "Cell surface receptors in lysophospholipid signaling." Semin Dev Biol 15: 457-465. Asakura, M., M. Kitakaze, S. Takashima and Y. Liao (2002). "Cardiac hypertrophy is inhibited by antagonism of ADAM-12 processing of HB-EGF: MMP inhibitors as a new therapy." Nat Med 8: 35-40. Bakken, A., C. Protack, E. Roztocil, S. Nicholl and M. Davies (2009). "Cell migration in response to the amino-terminal fragment of urokinase requires epidermal growth factor receptor activation through an ADAM-mediated mechanism." J Vasc Surg 49(5): 1296-1303. Barbolina, M., B. Adley, E. Ariztia, Y. Liu and M. Stack (2007). "Microenvironmental regulation of membrane type 1 matrix metalloproteinase activity in ovarian carcinoma cells via collagen-induced Egr-1 expression." J Biolo Chem 282: 4924- 4931. 257

Baron, V., G. De Gregorio, A. Krones-Herzig and T. Virolle (2003). "Inhibition of Egr-1 expression reverses transformation of prostate cancer cells in vitro and in vivo." Oncogene 22: 4194-4204. Barret, A. (1981). "Alpha 2-macroglobulin " Meth Enzymol 80: 737-754. Barret, T. and E. Bemditt (1987). "Sis (platelet-derived growth factor B chain) gene transcription levels are elevated in human atherosclerotic lesions compared to normal artey." Proc Natl Acad Sci USA 84: 1099-1103. Barret, T., C. Gadjusek, S. Schwartz, J. McDougall and E. Benditt (1984). "Expression of the sis gene by endothelial cells in culture and in vivo." Proc Natl Acad Sci USA 81: 6772-6774. Berasain, C., M. Perugorria, M. Latasa and J. Castillo (2009). "The epidermal growth factor receptor: a link between inflammation and liver cancer." Exp Biol Med 234: 713- 725. Berk, B. (2003). "Angiotensin Type 2 Receptor (AT2R): A challenging twin." Science 181: 1-3. Betsholtz, C., L. Karlsson and P. Lindahl (2004). "Developmental roles of platelet-derived growth factors." Bioessays 23: 26-45. Bhindi, R., L. Khachigian and H. Lowe (2006). "DNAzymes targeting the transcription factor Egr-1 reduce miocardial infarct size following ischemia-reperfusion in rats." JThromb and Haemost 4(7): 1479-1483. Billet, M., I. Adbeish, S. Alrokayan, A. Bennet, C. Marenah and D. White (1996). "Increased expression genes for platelet derived-growth factor in circulating mononuclear cells of hypercholesterolemic patients." Arterioscler Thromb Vasc Biol 16: 399-406. Bishayee, A., L. Beguinot and S. Bishayee (1999). "Phosphorylation of tyrosine 992, 1068 and 1086 is required for conformational change of the human epidermal growth factor receptor C-terminal tail." Molec Biol Cell 10: 525-536. Black, R., C. Rauch and C. Kozlosky (1997). "A metalloproteinase disintegrin that releases tumour-necrosis factor-alpha from cells." Nature 385: 729-733. Blaschke, F., D. Bruemmer and R. Law (2004). "Egr-1 is a major vascular pathogenic transcription factor in atherosclerosis and restenosis." Rev in Endoc & Metab Disor 5: 249-254. Bonner, J. (2004). "Regulation of PDGF and its receptors in fibrotic diseases." Cytok & Growt Fact Rev 15: 255-273. Boss, M., J. Mendelsohn, Y. Kim, J. Albanell, D. Fry and J. Baselga (1997). "PD153035, a tyrosine kinase inhibitor, prevents epidermal growth factor receptor activation and inhibits growth of cancer cells in a receptor number-dependent manner." Clin Cancer Res 3: 2099-2106. Buteau, J., S. Foisy, E. Joly and M. Prentki (2003). "Glucagon-like peptide 1 induces pancreatic beta-cell proliferation via transactivation of the epidermal growth factor receptor " Diabet 52: 124-132. Cahill, M., R. Janknecht and A. Nordheim (1996). "Signalling pathways: Jack of all cascades." Curren Biol 6(1): 16-19. Campbell, B. and J. A. Robertson (1981). "Effects of angiotensin II and vasopressin on human smooth muscle cells in vitro." Exp Mol Pathol 35: 265-276. 258

Campbell, G. and J. Campbell (1990). "The phenotypes of smooth muscle expressed in human atheroma." Ann N Y Acad Sci 598: 143-158. Campbell, J., S. Hughes, D. Gilbertson, T. Palmer, M. Holdren, A. Haran, M. Odell and R. Bauer (2005). "Platelet-derived growth factor C induces liver fibrosis, steatosis, and hepatocellular carcinoma." Proc Natl Acad Sci USA 102: 3389-3394. Canault, M., K. Peiretti, F. Kopp and B. Bornardo (2006). "The TNF alpha converting enzyme (TACE/ADAM17) is expressed in the atherosclerotic lesions of apolipoprotein E-deficient mice: possible contribution to elevated plasma levels of soluble TNF alpha receptors." Atheroscl 187: 82-91. Cao, R., E. Brakenhielm, X. Li, K. Pietras, J. Widenfalk, A. Ostman, U. Eriksson and Y. Cao (2002). "Angiogenesis stimulated by PDGF-CC, a novel member in the PDGF family, involves activation of the PDGFR-alphaalpha and -alphabeta receptors." FASEB 16: 1575-1583. Chakraborti, S., M. Mandal, S. Das, A. Mandal and T. Chakraborti (2003). "Regulation of matrix metalloproteinases." Molec Cell Biochem 253: 269-285. Chan, H., N. Smith, R. Hannah and T. Walter (2006). "Tackling the EGFR in pathological tissue remodelling." Pulmon Pharmacol Therap 19: 74-78. Chan, J., L. Prado-Lourenco, L. Khachigian, M. Bennet, M. Bartolo and M. Kavurma (2010). "TRAIL promotes VSMC proliferation and neointima formation in a FGF- 2-, Sp1 phosphorylation-, and NFkappaB-dependent manner." Circ Res 106(6): 1061-1071. Chang, C. and Z. Werb (2001). "The many faces of metalloproteases: cell growth, invasion, angiogenesis and metastasis." Trends Cell Biol 11(11): S37-43. Chen, A., J. Xu and A. Johnson (2006). "Curcumin inhibits human colon cancer cell growth by suppressing gene expression of epidermal growth factor receptor through reducing the activity of the transcription factor Egr-1." Oncogene 25: 278-287. Chen, X., T. Beers-Gibson, F. Robinson, L. Silvestro and G. Pearson (2001). "MAP Kinases." Chem Rev 101: 2449-2476. Chiu, I. and E. Reddy (1984). "Nucleotide sequence analysis identifies the human-cis proto-oncogene as a structural gene for platelet-derived growth factor." Cell 37(1): 123-129. Cho, A., J. Graves and M. Reidy (2000). "Mitogen-activated protein kinases mediate matrix metalloproteinses-9 expression in vascular smooth muscle cells." Arterioscler Thromb Vasc Biol 20: 2527-2532. Choi, S., M. Marazita, S. Hart, P. Sulima, L. Field and T. McHenry (2009). "The PDGF-C regulatory region SNP rs28999109 decreases promoter transcriptional activity and is associated with CL/P." Europ J of Hum Genet 17: 774-784. Chow, A., J. Cerna and R. Schulz (2007). "Acute actions and novel targets of matrix metalloproteinases in the heart and vasculature." Brit J of Pharmacol 152: 189-205. Chu, S. and J. Ferro (2005). "Sp1: Regulation of gene expression by phosphorylation." Gene 348: 1-11. Citri, A. and Y. Yarden (2006). "EGF-ERBB signaling: towards the system level." Nat Rev Mol Cell Biol 7: 505-516. Clark, I., T. Swingler, C. Sampieri and D. Edwards (2008). "The regulation of matrix metalloproteinases and their inhibitors." Intern J Biochem Cell Biol 5: 25-42. 259

Clowes, A., M. Reidy and M. Clowes (1983). "Kinetics of cellular proliferation after arterial injury. I. Smooth muscle growth in the absence of endothelium." Lab Invest 49: 327-333. Couillard, J., M. Demers, G. Lavoie and Y. St-Pierre (2006). "The role of DNA hypomethylation in the control of stromelysin gene expression." Biochem Biophys Res Commun 342: 1233-1239. Courey, A. and R. Tijian (1988). "Analysis of Sp1 in vivo reveals multiple transcriptionl domains, including a novel glutamine-rich activation motif." Cell 55: 887-898. Crawford, Y., I. Kasman, L. Yu, C. Zhong, X. Wu, Z. Modrusan, J. Kaminker and N. Ferrara (2009). "PDGF-C mediates the angiogenic and tumorigenic properties of fibroblasts associated with tumors refractor to anti-VEGF treatment." Cancer Cell 15: 21-34. Dalla-Favera, R. and R. Gallo (1982). "Chromosomal location of the human homolog (c- sis) of the simian sarcoma gene." Science 218: 686-688. Daub, K., C. Wallasch, A. Lankenau, A. Herlich and A. Ullrich (1997). "Signal characteristics of G protein -transactivated EGFR." EMBO J 16(23): 7032-7044. Day, F., L. Rafty, C. Chesterman and L. Khachigian (1999). "Angiotensin II (ATII)- inducible platelet derived growth factor A-chain gene expression is p42/44 extracellular signal-regulated kinase-1/2 and Egr-1-dependent and mediated via the ATII type 1 but not type 2 receptor." J Biol Chem 274(34): 23726-23733. Decker, S. and P. Harris (1989). "Effects of platelet-derived growth factor on phosphorylation of the epidermal growth factor receptor in human skin fibroblasts." J Biolo Chem 264(5): 9204-9209. Deguchi, J., M. Makuuchi, T. Nakaoka, T. Collins and Y. Takuwa (1999). "Angiotensin II stimulates platelet-derived growth factor-B chain expression in newborn rat vascular smooth muscle cells and neointimal cells through Ras, extracellular signal- regulated protein kinase, and c-Jun N-terminal protein kinase mechanisms." Circ Res 85(7): 565-574. Delbridge, G. and L. Khachigian (1997). "FGF-1-induced PDGF-A chain gene expression in vascular endothelial cells involves transcriptional activation by Egr-1." Circ Res 81: 282-288. DeMail, K., S. Godwin, S. Soltoff and A. Kaziauskas (1999). "Muliple role of Src in a PDGF-stimulated cell." Exper Cell Res 253: 271-279. Ding, H., X. Wu, H. Bostrom, I. Kim and N. Wong (2004). "A specific requirement for PDGF-C in palate formation and PDGFR-alpha signalling." Nat Genet 36: 1111- 1116. Doolittle, R. and M. Hunkapiller (1983). "Simian sarcoma virus onc gene, v-sis, is derived from the gene (or genes) encoding a platelet-derived growth factor." Science 221: 275-277. Downward, J. and M. Parker (1984). "Autophosphorylation sites on the epidermal growth factor receptor " Nature 311: 483-485. Durchdewald, M., P. Angel and J. Hess (2009). "The transcription factor fos: a janus-type regulator in health and disease." Histol Histopathol 24(11): 1451-1461. Dwivedi, D., G. Pino, A. Banh and D. Howchin (2006). "Matrix metalloproteinase inhibitors suppress transforming growth factor-beta-induced subcapsular cataract formation." Am J Pathol 168(1): 69-79. 260

Edwards, D., M. Handsley and C. Pennington (2008). "The ADAM metalloproteinases." Molec Aspec Medic 29: 258-289. Eguchi, S., P. Dempsey, G. Frank, E. Motley and T. Inagami (2001). "Activation of MAPKs by ATII in vascular smooth muscle cells: metalloproteinase dependent EGFR activation is required for activation of ERK and p38 MAPK but not JNK." J Biolo Chem 276: 7957-7962. Eitner, F., T. Ostendorf, M. Kretzel, C. Cohen and J. Floege (2002). "PDGF-C Expression in the developing and normal adult human kidney and in glomerular diseases." J Am Soc Nephrol 14: 1145-1153. Eriksson, A., A. Siegbahn, B. Westmark and C. Heldin (1992). "PDGF alpha and beta receptors activate unique and common signal transduction pathways." EMBO 11: 543-550. Fabunmi, R., A. Baker, E. Murray, R. Booth and A. Newby (1996). "Divergent regulation by growth factors and cytokines of 95 kDa and 72kDa gelatinases and tissue inhibitors or metalloproteinases-1, -2 and -3 in rabbit aortic smooth muscle cells." Biochem J 315 (Pt 1): 335-342. Fahmy, R., C. Dass, L. Sun, C. Chesterman and L. Khachigian (2003). "Transcription factor Egr-1 supports FGF-dependent angiogenesis during neovascularization and tumor growth." Nat Med 9(8): 1026-1032. Fahmy, R. and L. Khachigian (2002). "Antisense Egr-1 RNA driven by the CMV promoter is an inhibitor of vascular smooth muscle cell proliferation and regrowth after injury." J Cel Bioch 84: 575-582. Fahmy, R. and L. Khachigian (2006). "Suppression of growth factor expression and human vascular smooth muscle cell growth by small interfering RNA targeting Egr-1." J Cel Chem. Fang, L., Y. Yan, L. Komuves, S. Yonkovich, C. Sullivan, B. Stringer, P. Nurden and N. Giese (2004). "PDGF-C is a selective alfa platelet derived growth factor receptor agonist that is highly expressed in platelet alfa granules and vascular smooth muscle." Arter Thromb Vasc Biol 24: 787-792. Fears, C., J. Grammer and J. Stewart (2005). "Low-density lipoprotein receptor-related protein contributes to the angiogenic activity of thrombospondin-2 in a murine glioma model." Canc Res 65(20): 9338-9346. Fingleton, B. (2006). "Matrix metalloproteinases: roles in cancer and metastasis." Front Biosci(11). Fredrikson, L., H. Li and U. Eriksson (2004). "The PDGF family: four gene produtcs form five dimeric isoforms." Cytok & Growth Fac Rev 15: 197-204. Frid, M., E. Moiseeva and R. Stenmark (1994). "Multiple phenotypically distinct smooth muscle cell populations exist in the adult and developing bovine pulmonary arteial media in vivo." Circ Res 75(669-81). Fuentes, L., C. Reyes and J. Sarmiento (2008). "Vasopressin up-regulates the expression of growth-related immediate-early genes via two distinct EGFR transactivation pathways." Cel Signal 20: 1642-1650. Galis, Z. and J. Khatri (2002). "Matrix metalloproteinases in vascular remodeling and atherogenesis: the good, the bad and the ugly " Circ Res 90: 251-262. 261

Garton, K., P. Gough, J. Philalay, P. Willie and C. Blobel (2003). "Stimulated shedding of vascular cell adhesion molecule 1 (VCAM-1) is mediated by tumor necrosis factor- alpha-converting enzyme (ADAM 17)." J Biol Chem 278: 37459-37464. Giachelli, C., N. Bae, M. Almeida, D. Denhart, C. Alpers and S. Schwartz (1993). "Osteoponin expression is elevated during neointima formation in rat arteries and is a novel component of human atherosclerotic plaques." J Clin Invest 92: 1686-1696. Gibbons, J., D. Jones, T. Gadner, L. Goldstein, J. Moller and C. Yancy (2008). "The American Heart association's 2008 statement of principles for healthcare reform." Circ Res 2008(118): 2209-2218. Gilberston, D., M. Duff and J. West (2001). "Platelet-derived growth factor C (PDGF-C) a novel growth factor that binds to PDGF-apha and beta receptor." J Biolo Chem 276: 27406-27414. Glukhova, M., M. Frid and V. Koteliansky (1991). "Phenotypic changes of human aortic smooth muscle cells during development and in the vessel wall." Am J Pathol 261: 78-80. Gomez, D., D. Alfonso, H. Yoshiji and U. Thorgeirsson (1997). "Tissue inhibitors of metalloproteinases: structure, regulation and biological functions." Eur J Cell Biol 74: 111-122. Gonzales, P., A. Solomon and A. Miller (2004). "Inhibition of the tumor necrosis factor- alpha-converting enzyme by its pro domain." J Biolo Chem 279: 31683-31645. Gooz, M. (2010). "ADAM-17: The enzyme that does it all." Crit Rev Biochem Mol Biol 45(2): 146-169. Griendling, K., M. Ushio-Fukai, B. Lassegue and R. Alexander (1997). "Angiotensin II signalling in vascular smooth muscle cells: new concepts." Hypert 29: 366-373. Grun, K., M. Boyka, F. Bohmer and C. Leipner (2005). "Elevated expression of PDGF-C in coxsackievirus B3-induced chronic myocarditis." Europ Heart J 26: 728-739. Gschwind, A., S. Hart, O. Fisher and A. Ullrich (2003). "TACE cleavage of proamphiregulin regulates GPCR-induced proliferation and motility of cancer cells " EMBO 22: 2411-2421. Gschwind, A., E. Zwick, N. Prenzel, M. Leserer and A. Ullrich (2001). "Cell communication networks: epidermal growth factor receptor transactivation as the paradigm for inter-receptor signal transmission." Oncogene 20: 1594-1600. Gurjar, M., J. DeLeon, R. Sharma and R. Bhalla (2001). "Role of reactive oxygen species in IL-1 beta-stimulated sustained ERK activation and MMP-9 induction." Am J Physiol Heart Circ Physiol 281: H2568-2574. Gutkind, J. (1998). "Cell growth control by G-protein-coupled receptors: from signal transduction to signal integration." Oncogene 17: 1331-1342. Habib, A., H. Thorbergur, J. Gren, K. Stefansson and R. Ratan (1986). "The epidermal growth factor receptor associates with and recruits phosphatidylinositol 3-kinase to the platelet-derived growth factor beta receptor." J Biolo Chem 273: 6885-6891. Haendeler, J. and B. Berk (2000). "Angiotensin II mediated signal transduction : Important role of tyrosine kinases." Reg Pep 95: 1-7. Heldin, C., U. Eriksson and A. Ostman (2002). "New platelet-derived growth factor family of mitogens." Arch Biochem Biophys 398(2): 284-290. Heldin, C., A. Ostman and L. Ronnstrand (1998). "Signal transduction via platelet-derived growth factor receptors." Bioch and Byoph Acta 1378: F79-F113. 262

Heldin, C. and B. Wang (1999). "Mechanisms of action and in vivo role of platelet-derived growth factor." Physiol Rev 79(4): 1284-1300. Herbert, J., C. Delisee and F. Dol (1994). "Effect of SR 47436, a novel angiotensin II AT1 receptor antagonist, on human vascular smooth muscle cells in vitro." Eur J Pharmacol 251: 143-150. Higashiyama, S., H. Iwabuki, C. Morimoto, M. Hieda and N. Matsushita (2008). "Membrane-anchored growth factors, the epidermal growth factor family: beyond receptor ligands." Canc Scien 99(2): 214-220. Hoegy, S., H. Oh, M. Corcoran and W. Stetler-Stevenson (2001). "Tissue inhibitor of metalloproteinase-2 (TIMP-2) supresses TKR-growth factor signaling independent of metalloproteinase inhibition." J Biol Chem 276: 3203-3214. Hollera, W., R. Galardy, G. W, D. Levy, P. Tang and P. Elias (1997). "Matrix metalloproteinase inhibitors reduce phorbol ester-induced cutaneous inflammation and hyperplasia." Arch Dermatol Res 289: 138-144. Horiuchi, M. (1997). "Angiotensin type 2 receptor dephosphorylates Bcl-2 by activating mitogen-activated protein kinsae phosphatase-1 and induces apoptosis." J Biol Chem 272: 19022-19026. Hou, X., A. Kumar, C. Lee and B. Wang (2010). "PDGF-CC blockade inhibits pathological angiogenesis by acting on multiple cellular and molecular targets." Proc Natl Acad Sci USA 107: 12216-12221. Huang, X. (1996). "Mitogen-activated protein kinases in rat brain neuronal cultures are activated by angiotensin II type 1 receptors and inhibited by angiotensin II type 2 receptors." J Biol Chem 271: 15635-15641. Huez, I., L. Creancier, S. Audigier, M. C. Gensac, A. C. Prats and H. Prats (1998). "Two independent internal ribosome entry sites are involved in translation initiation of vascular endothelial growth factor mRNA." Mol Cell Biol 18(11): 6178-6190. Hughes, S. (1993). "Expression of basic and acid fibroblast growth factors and their receptor in normal and atherosclerotic human arteries." Cardiovas Res 27: 1214- 1219. Hurtado, M., J. Lozano and E. Castellanos (2007). "Activation of the epidermal growth factor signalling pathway by tissue plasminogen activator in pancreas cancer cells." Gut 56: 1266-1274. Igura, T., J. Kawata and J. Miyagawa (1996). "Expression of heparin-binding epidermal growth factor-like in neointimal cells induced by balloon injury in rat carotid arteries." Arterioscler Thromb Vasc Biol 16: 1639-1648. Ingram, J., A. Antao-Menezes, J. Mangum, O. Lyght, P. Lee, J. Elias and J. Bonner (2006). "Opposing actions of Stat1 and Stat6 on IL-13-induced up-regulation of early growth response-1 and platelet-derived growth factor ligands in pulmonary fibroblasts." J Immunol 15(177): 4141-4148. Irvine, C., S. George, E. Shefield, J. Johnsosn and P. Lamonton (2000). "The association of platelet-derived growth factor receptor expression, plaque morphology and histological features with symptons in carotid atherosclerosis." Cardiovasc Surg 8: 121-129. Ishizaka, N., G. Matsuzaki, K. Saito, E. Noiri, I. Mori and R. Nagai (2006). "Expression and localization of PDGF-B, PDGF-D and PDGF receptor in kidney of angiotensin II-infused rat." Lab Invest 86(12): 1285-1292. 263

Iwami, K., N. Ashizawa, Y. S. Do, K. Graf and W. A. Hsueh (1996). "Comparison of angiotensin II with other growth factors on Egr-1 and matrix gene expression in cardiac fibroblasts." Am J Physiol 270(6 Pt 2): H2100-2107. Jawien, A., D. Bowen-Pope, V. Linder, S. Schwartz and A. Clowes (1992). "Platelet- derived growth factor promotes smooth muscle cell migration and intimal thickening in a rat model of balloon angioplasty." J Clin Invest 89(2): 507-511. Kamikura, M., F. Bea, T. Akizawa and H. Katus (2004). "Platelet-derived growth factor induces tissue factor expression in vascular smooth muscle cells via activation of Egr-1." Hypert 44: 944-951. Karvinen, H., J. Rutanen, O. Leppanen, R. Lach, A. Levonen, U. Eriksson and T. Yia- Herttualla (2009). "PDGF-C and -D and their receptors PDGFR-alpha and PDGFR- beta in atherosclerotic human arteries." Eur J Clin Inves 39(4): 320-327. Kato, M., T. Inazu and Y. Kawai (2003). "Amphiregulin is a potent mitogen for the vascular smooth cell line, A7R5." Biochem Biophys Res Commun 301: 1109-1115. Kaufman, K. and G. Thiel (2001). "Epidermal growth factor and platelet-derived growth factor induce expression of Egr-1, a zinc finger transcription factor in human malignant glioma cells." J Neurolo Scien 189: 83-91. Kavurma, M. and L. Khachigian (2003). "ERK, JNK, and p38 MAP kinases differentially regulate proliferation and migration of phenotypically distinct smooth muscle cell subtypes." J Cel Biochem 89: 289-300. Kawagishi, J., T. Kumabe, T. Yoshimoto and T. Tamamoto (1995). "Structure, organisation and transcription units of the human alpha-platelet derived growth factor receptor gene." Genom 30: 224-232. Kazlauskas, A., D. Durden and A. Cooper (1991). "Functions of the major tyrosine phosphorylation site of the PDGF receptor beta subunit " Cell Reg 2: 413-425. Keates, S., A. C. Keates, S. Nath, R. M. Peek, Jr. and C. P. Kelly (2005). "Transactivation of the epidermal growth factor receptor by cag+ Helicobacter pylori induces upregulation of the early growth response gene Egr-1 in gastric epithelial cells." Gut 54(10): 1363-1369. Kessenbrock, K., V. Plaks and Z. Web (2010). "Matrix metalloproteinases: regulators of the tumor microenvironment." Cell 141: 52-67. Khachigian, L. (2000). "Catalytic DNA as a therapeutic agents and molecular tools to disect biological functions." J Clin Inves 106: 1189-1195. Khachigian, L. (2004). "DNAzymes as molecular agents that manipulate Egr-1 gene expression." Bioch Pharmac 68(6): 1023-1925. Khachigian, L. (2006). "Early growth response-1 in cardiovascular pathobiology." Circ Res 98(2): 186-191. Khachigian, L. and T. Collins (1997). "Inducible expression of Egr-1-dependent genes." Circ Res 81: 457-461. Khachigian, L. and T. Collins (1998). "Early growth response factor 1: a pleiotropic mediator of inducible gene expression." J Mol Med 76(9): 613-616. Khachigian, L., U. Fries, M. Benz, D. Bonthron and T. Collins (1994). "Novel cis-acting elements in the human platelet-derived growth factor B-chain core promoter mediate gene expression in cultured vascular endothelial cells." J Biol Chem 269: 22647-22656. 264

Khachigian, L., V. Linder, A. Williams and T. Colllins (1996). "Egr-1-induced endothelial gene expression: a common theme in vascular injury." Science 271: 1427-1431. Khachigian, L., Y. Takuwa and T. Collins (2000). "Mechanisms of angiotensin-II-induced platelet derived growth factor gene expression." Molec Cel Biochem 212: 183-186. Khachigian, L., A. Williams and C. T (1995). "Interplay of Sp1 and Egr-1 in the proximal platelet derived growth factor A-chain promoter in cultured vascular endothelial cells." J Biol Chem 270(46): 26679-27686. Kocker, O., O. Skalli, W. Bloom and G. Gabbiani (1984). Cytoskeleton of rat aortic smooth muscle cells: normal conditions and experimental intimal thickening. Lab Invest. 50: 645-652. Kohno, M., K. Ohmori and S. Nozaki (2000). "Effects of valsartan on angiotensin II- induced migration of human coronary artery smooth muscle cells." Hypertens Res 23: 677-681. Kong, W., J. Kang and A. Walding (2010). "Interleukin-1beta induces migration of rat arterial smooth muscle cells through a mechanism involving increased matrix metalloproteinase-2 activity." J Surg Res 5: 145-150. Krettet, A., G. Fager, P. Jerberg and F. Lusting (1997). "Quantitation of platelet-derived growth factor receptors in human arterial smooth muscle cells in vitro." Arterioscler Thromb Vasc Biol 81: 807-869. Kuma, S., E. Oki, T. Onohara, K. Komori and Y. Maehara (2007). "Angiotensin II-induced growth of vascular smooth muscle cells is associated with modulation of cell surface area and platelet-derived growth factor receptor expression." Clin Exp Pharm Phys 34: 153-160. Lamb, D., H. Modjtahedi, N. Plant and G. Ferns (2004). "EGF mediates monocyte chemotaxis and macrophage proliferation and EGFR is expressed in atherosclerotic plaques." Atheroscl 176: 21-26. Lautrette, A., S. Li, R. Alili and S. Sunnaborg (2005). "Angiotensin II and EGFR cross-talk in chronic kidney diseases: a new therapeutic approach." Nat Med 11: 867-874. Lee, K. A., T. Y. Hai, L. SivaRaman, B. Thimmappaya, H. C. Hurst, N. C. Jones and M. R. Green (1987). "A cellular protein, activating transcription factor, activates transcription of multiple E1A-inducible adenovirus early promoters." Proc Natl Acad Sci U S A 84(23): 8355-8359. Lei, H. and A. Kazlauskas (2008). "Focus on molecules: platelet-derived growth factor C: PDGF-C." Exp Eye Res 86: 711-712. Lemarie, C. and E. Schiffrin (2010). "The angiotensin II type 2 receptor in cardiovascular disease." J Reni-Angioten-Aldoster Syst 11(1): 19-31. Lemire, J., C. Covin, S. White, C. Giachelli and S. Schwatz (1994). "Characterization of cloned aortic smooth muscle cells from young rats." Am J Pathol 144: 1068-1081. Lemire, J., S. Potter-Perigo, K. Hall, T. Wight and S. Schwartz (1996). "Distinct rat aortic smooth muscle cells differ in versican/PG-M expression." Arterioscler Thromb Vasc Biol 16: 821-829. Leveen, P., M. Pekny and S. Gebre-Medhin (1994). "Mice deficient for PDGF-B show renal, cardiovascular and haematological abnormalities." Genes Dev 8: 1875-1887. Li, A. and C. Glass (2002). "The macrophage foam cells as a target for therapeutic intervention." Nat Med 8: 1235-1242. 265

Li, J., Y. Kim and P. Bertics (2000). "Platelet-derived growth factor-stimulated migration of murine fibroblasts is associated with epidermal growth factor receptor expression and tyrosine phosphorylation " J Biol Chem 275(4): 2951-2958. Li, W. and J. Miller (2000). "Enhanced H202-induced cytotoxicity in epithelioid smooth muscle cells: implications for neointimal regression." Arterioscler Thromb Vasc Biol 20: 1473-1479. Li, X. and U. Eriksson (2003). "Novel PDGF family members:PDGF-C and PDGF-D." Cytok & Grow Fac Rev 14: 91-98. Li, X., J. Jiang and S. Frank (2010). "Synergy in ERK activation by citokine receptors and tyrosine kinase growth factor receptors." Cell Signal 11: 148-156. Li, X., Y. Meng, M. Huang, X. Zhang and Z. Zhang (2008). "Angiotensin II stimulates platelet-derived growth factor-B expression in hepatic stellate cells by activating Egr-1." Nan Fang Yi Ke Da Xue Xue Bao 28(6): 963-967. Li, X., A. Ponten, K. Aase, L. Karisson, H. Li, P. Soriano, K. Alitalo, A. Ostmans and U. Eriksson (2000). "PDGF-C is a new protease-activated ligand for the PDGF alfa receptor." Nat Cell Biol 2: 302-309. Li, X., M. Tjwa and L. Moons (2005). "Revascularization of ischemic tissues by PDGF-CC via effects on endothelial cells and their progenitors." J Clin Invest 115: 118-127. Libby, P. and M. Aiwaka (2002). "Stabilization of athersclerotic plaques: new mechanisms and clinical targets." Nat Med 8: 1257-1262. Lieu, C. and S. Kopetz (2009). "The Src family of protein tyrosine kinases: a new and promising target for colorectal cancer therapy." Clin Colorectal Canc 9(2): 89-94. Lin, J. and W. Leonard (1997). "The immediate-early gene product Egr-1 regulates the human interleukin-2-receptor beta-chain promoter through noncanonical Egr-1 and Sp1 binding sites." Mol Cell Biol 17: 3714-3722. Lindner, V. and M. Reidy (1995). "Platelet-derived growth factor ligand and receptor expression by large vessel endothelium in vivo." Amer J Path 146(6): 1488-1497. Ling, S., A. Dai, Y. Ma, E. Wilson, K. Chatterjee, H. Ives and K. Sudhir (1999). "Matrix- dependent gene expression of Egr-1 and PDGF-A regulates angiotensin II-induced proliferation in human vascular smooth muscle cells." Hyperten 34(5): 1141-1146. Liu, L., J. Tsai and W. Aird (2000). "Egr-1 gene is induced by the systemic administration of the vascular endothelial growth factor and the epidermal growth factor." Blood 96: 1772-1781. Liu, M., M. Eyries, C. Zhang, F. Santiago and L. Khachigian (2006). "Inducible platelet- derived growth factor D-chain expression by angiotensin II and hydrogen peroxide involves transcriptional regulation by Ets-1 and Sp1." Blood 107(6): 2322-2329. Liu, P. and R. Anderson (1999). "Spatial organization of EGF receptor transmodulation by PDGF." Biochem and Biophys Res Comm 261: 695-700. Liu, P., Y. Ying and Y. Ko (1996). "Localization of platelet-derived growth factor- stimulated phosphorylation cascade to caveolae " J Biol Chem 271: 10299-10303. Lowe, H., C. Chesterman and L. Khachigian (2002). "Catalytic antisense DNA molecules targeting Egr-1 inhibit neointima formation following permanent ligation of rat carotid arteries." Thromb Haemost 87: 134-140. Lowe, H., R. Fahmy, M. Kavurma, A. Baker, C. Chesterman and L. Khachigian (2001). "Catalytic oligodeoxynucleotides define a key regulatory role of early growth 266

response factor-1 in the porcine model of coronary in-stent restenosis." Circ Res 89: 670-677. Majesky, M., E. Benditt and S. Schwartz (1988). "Expression and developmental control of platelet-derived growth factor A-chain and B-chain/Sis genes in rat aortic smooth muscle cells." Proc Natl Acad Sci USA 85: 1524-1528. Majesky, M., M. Daemen and S. Schwartz (1990). "Alpha 1-adrenergic stimilation of platelet-derived growth factor-A chain gene expression in rat aorta " J Biolo Chem 265(2): 1082-1088. Majesky, M., C. Giachelli, M. Reidy and S. Schwartz (1992). "Rat carotid neointimal smooth muscle cells re-express a developmentally regulated mRNA phenotype during repair of arterial injury." Circ Res 71: 759-768. Martinez-Gonzalez, J., M. Berrozpe, O. Varela and L. Badimon (2002). "Heterogeneity of smooth muscle cells in advanced human atherosclerotic plaques: intimal smooth muscle cells expressing a fibroblast protein are highly activated by platelet-released products." Eur J Clin Invest 31: 939-949. Masataka, S. and D. Fukuda (2010). "Crucial role of renin-angiotensin system in the pathogenesis of atherosclerosis." J Med Invest 57: 12-15. Massague, J. and A. Pandiella (1993). "Membrane-anchored growth factors." Annu Rev Biochem 62: 515-541. McCafrrey, T., C. Fu, B. Du, S. Eksnar, C. Kent, E. Silverman and T. Collins (2000). "High-level expression of Egr-1 and Egr-1 inducible genes in mouse and human atherosclerosis." J Clin Inves 105(105): 653-662. McCole, D., S. Keely, R. Coffet and K. Barrett (2002). "Transactivation of the epidermal growth factor receptor in colonic epithelial cells by carbachol requires extracellular release of transforming growth factor-alpha." J Biolo Chem 277: 42603-42612. McDonald, O. and G. Owens (2007). "Programming smooth muscle cell plasticity with chromatin dynamics." Circ Res 100: 1428-1441. Mehta, P. and K. Griendling (2007). "Angiotensin II cell signaling: physiological and pathological effects in the cardiovascular system." Am J Pathol 292: C82-C97. Mendelson, K., S. Swendeman, P. Saftig and C. Bobel (2010). "Stimulation of PDGFRbeta activates ADAM17 and promotes metalloproteinase-dependent cross-talk between the PDGFRbeta and EGFR signaling pathways." J Biol Chem 285(32): 25024- 25032. Midgley, V. and L. Khachigian (2004). "Fibroblast growth factor-2 induction of platelet- derived growth factor-C chain transcription in vascular smooth muscle cells is ERK-dependent but not JNK-dependent and mediated Egr-1." J Biol Chem 279(24): 40289-40295. Mifune, M., H. Ohtsu, H. Suzuki, H. Nakashima, E. Brailoiu, N. Dun, G. Frank, T. Inagami, S. Higashiyama, W. Thomas, A. Eckhart, P. Dempsey and S. Eguchi (2005). " G-protein coupling and second messenger generation are indispensable for metalloprotease-dependent, heparin-binding epidermal growth factor shedding through angiotensin II type-1 receptor." J Biol Chem 280(8): 26592-26599. Milanini-Mongiat, J., J. Pouyssegur and G. Pages (2002). "Identification of two Sp1 phosphorylation sites for p42/p44 mitogen-activated protein kinases." J Biolo Chem 277: 20631-20639. 267

Milenkovic, I., M. Weick, P. Wiedemam, A. Reichenback and A. Bringmann (2003). "P2Y receptor-mediated stimulation of Muller glial cell DNA synthesis: dependence on EGF and PDGF receptor transactivation." Invest Ophtalm Vis Scie 44(3): 1211- 1220. Millette, E., B. Rauch, O. Defawe, R. Knagy, G. Daum and A. Clowes (2005). "Platelet- derived growth factor-BB-induced human smooth muscle cell proliferation depends on basic FGF release and FGFR-1 activation." Circ Res 96: 172-179. Minc, E., P. de Coppet, P. Masson, L. Thiery, S. Dutertre, M. Amor-Gueeret and C. Jaulin (1999). "The human copper-zinc superoxide dismutase gene (SOD1) proximal promoter is regulated by Sp1, Egr-1, and WT1 via non-canonical binding sites." J Biolo Chem 247(1): 503-509. Mineo, C., G. James, E. Smart and R. Andreson (1996). "Localization of platelet-derived growth factor-stimulated Ras/Raf-1 interaction to caveolae membrane." J Biolo Chem 271: 11930-11935. Mitsumata, M., S. Gamou, N. Shimizu and Y. Yoshida (1994). "Response of atherosclerotic intimal smooth muscle cells to epidermal growth factor in vitro " Arterioscler Thromb Vasc Biol 14: 1364-1371. Miyagawa, J., S. Higashiyama and S. Kawata (1995). "Localization of HB-EGF-like growth factor in the smooth muscle cells and macrophages in atherosclerotic plaques." J Clin Invest 95: 404-411. Montagut, C. and J. Settleman (2009). "Targeting the RAF-MEK-ERK pathway in cancer therapy." Canc Lett 283: 125-134. Moss, N. and E. Benditt (1970). "Spontaneous and experimentally induced arterial lessions. An ultrastructrural survey of the normal chicken aorta." Lab Invest 22: 166-183. Murate, T., K. Yamashita and H. Ohashi (1993). "Erytroid potentiating activity of tissue inhibitor of MMPs on the differentiation of erythropoietin-responsive mouse erythroleukemia cell line, ELM-I-1-3, is closely related to its cell growth potentiating activity " Exp Hematol 21(1): 169-176. Murillo, M., I. Carmona-Cuenca and G. Del Castillo (2007). "Activation of NADPH oxidase by transforming growth factor-beta in hepatocytes mediates up-regulation of epidermal growth factor receptor ligands through a Nuclear factor-kappa-B- dependent mechanism." J Biolo Chem 405: 251-259. Murillo, M., I. Carmona-Cuenca and A. Sanchez (2005). "Involvement of EGFR and c-Src in the survival signals induced by TGF-beta 1 in hepatocytes." Oncog 24: 4580- 4587. Nabel, E., Z. Yang, S. Liptay, H. San, D. Gordon, G. Haudenschild and E. Nabel (1993). "Recombinant platelet-derived growth factor B gene expression in porcine arteries induce intimal hyperplasia in vivo." J Clin Invest 91: 1822-1829. Nakajima, M., H. Hutchinson, M. Fujinaga, W. Hayashida, R. Mirishita, L. Zhang, M. Horiouchi, R. Pratt and V. Dzau (1995). "The angiotensin II type 2 (AT2) receptor antagonizes the growth effects of the AT1 receptor: gain-of-function study using gene transfer." Proc Nat Acad Sci USA 92: 10663-10667. Nakajima, M., H. Hutchison and M. Fuginaga ( 1995). "The angiotensin II type 2 (AT2R) receptor antagonizes the growth effects of the AT1R: gain-of-function study using gene transfer." Proc Natl Acad Sci USA 92: 10663-10667. 268

Nanney, L., C. Stoscheck and L. King (1988). "Characterization of binding and receptors for epidermal growth factor in smooth muscle." Cell Tissue Res 254. Newby, A. (2000). "An overview of the vascular response to injury: a tribute to the late Russell Ross." Toxicol Lett 112-113: 519-529. Newby, A. (2005). "Dual role of matrix metalloproteinases: matrixins in intimal thickening and atherosclerosis plaque rupture." Physiol Rev 85: 1-31. Newby, A. (2006). "Matrix metalloproteinase regulate migration, proliferation and death of vascular smooth muscle cells by degrading matrix and non-matrix substrates." Cardiovasc Res 69(3): 614-624. Newby, A. and A. Zaltsman (2000). "Molecular mechanisms in intimal hyperplasia." J Pathol 190: 300-309. NHF (2005). "Heart, stroke and vascular diseases. Australian facts 2004." National heart foundation 2005. Ning, W., Y. Dong, J. Sun, C. Li and M. Matthay (2007). "Cigarrette smokes stimulates matrix metalloproteinase-2 activity via Egr-1 in human lung fibroblasts." Am J Respir Cell Mol 36: 480-490. Nomoto, A., S. Mutoh, H. Hagihara and I. Yamaguchi (1988). "Smooth muscle cell migration induced by inflammatory cell products and its inhibition by a potent calcium antagonist, nilvapidine." Atheroscl 72: 213-219. Nouet, S. and C. Nahmias (2000). "Signal transduction from the angiotensin II AT2 receptor." TEM 11: 1-6. Ohtani, K., K. Egashira, K. Hiasa, Q. Zhao, S. Kitamoto, M. Ishibashi, M. Usui, S. Inoue, Y. Yonemitsu, K. Sueishi, M. Sata, M. Shibuya and K. Sunagawa (2004). "Blockade of vascular endothelial growth factor suppresses experimental restenosis after intraluminal injury by inhibiting recruitment of monocyte lineage cells." Circ 110(16): 2444-2452. Ohtani, K., K. Egashira, M. Usui, K. Hiasa, Q. Zhao, M. Aoki and A. Takeshita (2004). "Inhibition of neointimal hyperplasia after balloon injry by cis-element "decoy" of Egr-1 in hypercholesterolemic rabbits." Gen Ther 11: 126-132. Ohtsu, H., P. Dempsey and S. Eguchi (2006). "ADAMs as mediators of EGFR transactivation by G protein-coupled receptors." Am J Physiol Cell Physiol 291: C1-C10. Olson, N., S. Chao, V. Lindner and M. Reidy (1992). "Intimal smooth muscle cell proliferation after balloon catheter injury." Am J Path 140: 1017-1023. Ostman, A. (2004). "PDGF receptors-mediators of autocrine tumor growth and regulator of tumor vasculature and stroma." Cytok & Growth Fact Rev 15: 275-286. Owens, G., M. Kumar and B. Wamhoff (2004). "Molecular regulation of vascular smooth muscle cell differentation in develpment and disease." Physiol Rev 84: 767-801. Page-McCaw, A., A. Ewald and Z. Werb (2007). "Matrix metalloproteinases and the regulation of tissue remodelling." Nat Rev Mol Cell Biol 8: 221-233. Pahara, J., H. Shi, X. Chen and Z. Wang (2010). "Dimerization drives PDGF receptor endocytosis through a c-terminal hydrophobic motif shared by EGF receptor." Experim Cell Res 316: 2237-2250. Paxton, W. (1994). "The angiotensin II AT1 receptor is tyrosine and serine phosphorylated and can serve as a substrate for the src family of tyrosine kinases." Biochem Biophys Res Commun 200(1): 260-267. 269

Pender, S., J. Quin, I. Sanderson and T. Macdonald (2000). "Butyrate up-regulates stromesylin-1 production by intestinal mesenchymal cells." Am J Physiol Gastrointest Liver Physiol 279: G918-G924. Peters, S., B. Gotting, M. Trummel, H. Rust and A. Brattstrom (2001). "Valsartan for prevention of restenosis after stenting of type B2/C lesions: the VAL-PREST trial." J Invasive Cardiol 13: 93-97. Pierce, K., A. Tohgo, S. Ahn and M. Field (2001). "Epidermal growth factor receptor EGFR-dependent ERK activation by G-protein coupled receptors: a co-culture system for identifying intermediate upstream and downstream of heparin binding- EGF shedding " J Biolo Chem 276: 23155-23160. Pisarev, A. V., V. G. Kolupaeva, V. P. Pisareva, W. C. Merrick, C. U. Hellen and T. V. Pestova (2006). "Specific functional interactions of nucleotides at key -3 and +4 positions flanking the initiation codon with components of the mammalian 48S translation initiation complex." Genes Dev 20(5): 624-636. Pompili, V., D. Gordon, H. San, Z. Yang, D. Muller, G. Nabel and E. Nabel (1995). "Expression and function of a recombinant PDGF B gene in porcine arteries." Arterioscl Thromb Vasc Biol 15(2254-64). Poncet, N., J. Guillaume and G. Mouchiroud (2010). "Epidermal growth factor receptor transactivation is implicated in IL-6-induced proliferation and ERK1/2 activation in non-transformed prostate epithelial cells." Cel Signal 23: 572-578. Pontem, A., X. Li, P. Thoren, K. Aase, A. Ostman and U. Eriksson (2003). "Transgenic overexpression of platelet -derived growth factor-C in the mouse heart induces cardiac fibrosis, hypertrophy and dilated cardiomyopathy." Am J of Pathol 163(2): 673-682. Prenzel, N., E. Zwich, L. Daub and M. Leserer (1999). "EGFR transactivation by GPCR requires metalloproteinase cleavage of proHB-EGF." Nat 402(6764): 884-888. Qin, H., Y. Sun and E. Benveniste (1999). "The transcription factors Sp-1, Sp-3 and AP-2 are required for constitutive matrix metalloproteinase-2 gene expression in astroglioma cells." J Biolo Chem 274: 29130-29137. Raab, G. and M. Klagsbrun (1997). "Heparin-binding EGF-like growth factor." Biochem Biophys Acta 1333: F179-199. Rafty, L. and L. Khachigian (1997). "Zinc finger transcription factors mediate high constitutive platelet-derived growth factor-B expression in smooth muscle cells derived from the aortae of newborn rats." J Biolo Chem 273(10): 5758-5764. Rafty, L. and L. Khachigian (2001). "Sp1 phosphorylation regulates inducible expression of platelet-derived growth factor B-chain gene via atypical protein kinase C-zeta." Nucl Acid Res 29: 1027-1033. Raines, E. (2004). "PDGF and cardiovascular disease." Cytok & Growth Fact Rev 15: 593- 599. Rajagopalan, S., S. Kurz, T. Munzel and D. Harrison (1996). "Angiotensin II-mediated hypertension in the rat increases vascular superoxide production via membrane NADH/NADPH oxidase activation." J Clin Invest 97: 1916-1923. Raychowdhury, R., G. Schafer, J. Fleming, S. Rosewicz, B. Wiedenmann, T. Wang and M. Hocker (2002). "Interaction of early growth response protein 1 (Egr-1), specificity protein 1 (Sp1), and cyclic adenosise 3'5'-monophosphate response element binding 270

protein (CREB) at the proximal response element is critical for grastrin-dependent activation of the chromagranin A promoter." Mol Endocrinol 16(12): 2802-2818. Reape, T., V. Wilson and J. Kanckzxler (1997). "Detection and cellular localization of heparin-binding epidermal growth factor-like mRNA and protein in human atherosclerotic tissue " J Mol Cell Cardiol 29: 1639-1648. Reddy, E., K. Ramana, S. Srivastava, A. Bhatnagar and S. Srivastava (2009). "Aldose reductase regulates high glucose-induced ectodomain shedding of tumor necrosis factor (TNF)-alpha via protein kinase C-delta and TNF-alpha converting enzyme in vascular smooth muscle cells." Endrocrinol 150: 63-74. Reigstad, L., H. Sandes, O. Fkuge, O. Bruland, A. Muga, A. Martinez and J. Lillehaug (2003). "Platelet-derived growth factor (PDGF)-C, a PDGF family member with vascular endothelial growth factor-like structure." J Biol Chem 278: 17114-17120. Reigstad, L., L. V. Jan and J. Lillehaug (2005). "Structural and functional specificities of PDGF-C and PDFG-D, the novel members of the platelet-derived growth factors family." FEBS J 272: 5723-5741. Reuben, P. and H. Cheung (2006). "Regulation of MMPs gene expression by protein kinases." Front Biosci 11: 1199-1215. Roach, H., N. Yamada, K. Cheung and S. Killey (2005). "Association between the abnormal expression of matrix-degrading enzymes by human osteoarthritic chondrocytes and demethylation of specific CpG sites in the promoter regions." Arthritis Rheum 52: 3110-3124. Robson, M., L. Philips, A. Thomason and G. Pierce (1992). "Platelet-derived growth factor BB for the treatment of chronic pressure ulcers." Lancet 339: 23-25. Roelle, S., R. Grosse, A. Aigner, H. Krell, F. Czubayko and T. Gudermann (2003). "Matrix metalloproteinses 2 and 9 mediate epidermal growth factor receptor transactivation by gonadotropin-releasin hormone." J Biolo Chem 278: 47307-47318. Ross, R. (1986). "The pathogenesis of atherosclerosis: an update." N England J Med 134: 488-500. Ross, R. (1993). "The pathogenesis of atherosclerosis: a perspective for the 1990s " Nat Cell Biol 362: 801-809. Ross, R. (1995). "Cell biology of atherosclerosis." Annu Rev Physiol 57: 791-804. Ross, R. (1999). "Atherosclerosis: an inflammatory disease." N England J Med 340: 115- 126. Ross, R. and A. Glomset (1973). "Atherosclerosis and the arterial smooth muscle cell: Proliferation of smooth muscle is a key event in the genesis of the lesions of atherosclerosis." Scien 180: 1332-1339. Ross, R., J. Masuda, W. Raines, S. Katsuda, M. Sasahara, L. Malden, H. Masuki and H. Sato (1990). "Localization of PDGF-B protein in macrophages in all phases of atherogenesis." Scien 248: 1009-1012. Roudabush, F., K. Pierce and S. Mausle (2000). "Transactivation of the EGF receptor mediates IGF-1 stimulated Shc phosphorylation and ERK1/2 activation in COS-7 cells." J Biolo Chem 275: 22583-22589. Roztocil, E., S. Nicholl and M. Davies (2008). "Insulin-induced epidermal growth factor activation in vascular smooth muscle cells is ADAM-dependent." Surg 144(2): 245- 251. 271

Ruiz-Ortega, M., O. Lorenzo and J. Egido (2000). "Angiotensin II activates nuclear transcription factor kappa B through AT1 and AT2 in vascular smooth muscle cells: molecular mechanisms." Circ Res 86: 1266-1272. Rupprecht12, H., S. Vikas, J. Lacy, B. Sterzel and L. Coleman (1993). "PDGF-induced Egr-1 expression in rat mesangial cells is mediated through upstream serum response elements." J Cel Chem: F351-F360. Rupprecht, H. S., V; Lacy, J; Sterzel, B; Coleman, D (1993). "PDGF-induced Egr-1 expression in rat mesangial cells is mediate through upstream serum response elements." Am J Pathol 265: F351-F360. Sachinidis, A., P. Weisser, Y. Ko and K. Schulte (1992). "Angiotensin II induces formation of the Egr-1 protein in rat vascular smooth muscle cells." FEBS Lett 23(313 (2)): 109-112. Saito, S., G. Frank, E. Motley, P. Dempsey, H. Utsunomiya, T. Inagami and S. Eguchi (2002). "Metalloprotease inhibitor blocks angiotensin II-induced migration through inhibition of epidermal growth factor receptor transactivation." Biochem and Byophy Res Commun 294: 1023-1029. Saito, Y. and B. Berk (2001). "Transactivation: a novel signaling pathway from Angiotensin II to tyrosine kinase receptors " J Mol Cell Cardiol 33: 3-7. Saito, Y., J. Haendeler, Y. Hojo, K. Yamamoto and B. Berk (2001). "Receptor heterodimerization: essential mechanism for platelet-derived growth factor-induced epidermal growth factor receptor transactivation." Mol Cell Biol 21(19): 6387-6394. Sanderson, M., P. Dempsey and A. Dunbar (2006). "Control of ErbB signaling through metalloprotease mediated ectodomain shedding of EGF-like factors." Growth Fact 24(2): 121-136. Sano, H., T. Sudo, M. Yokode and T. Murayama (2001). "Functional blockade of platelet- derived growth factor receptor beta but not of receptor alpha prevents vascular smooth muscle cell accumulation in fibrous cap lesions in apolipoprotein E- deficient mice." Circ 103(24): 2955-2960. Santiago, F. and L. Khachigian (2001). "Nucleic acids based strategies as potential therapeutic tools: mechanistic considerations and implications to restenosis." J of Molec Med 79: 695-706. Santiago, F., H. Lowe, F. Day, C. Chesterman and L. Khachigian (1999). "Early growth response factor-1 induction by injury is triggered by release and paracrine activation by fibroblast growth factor-2." Am J Pathol 154(3): 937-944. Santiago, F., H. Lowe, M. Kavurma, C. Chesterman, A. Baker and L. Khachigian (1999). "New DNA enzyme targeting Egr-1 mRNA inhibits vascular smooth muscle proliferation and regrowth after injury." Nat Med 5(11): 1264-1269. Santiskulvong, C. and E. Rozengurt (2003). "Galardin GM6001, a broad-spectrum matrix metalloproteinase inhibitor blocks bombesin- and LPA-induced EGFR transactivation and DNA synthesis in Rat-1 cells." Experim Cell Res 290: 437-446. Sariahmetoglu, M., B. Crawford, H. Leon and J. Sawicka (2007). "Regulation of matrix metalloproteinase-2 (MMP-2) activity by phosphorylation." FASEB. Satoh, M., Y. Ishikawa, T. Itoh, Y. Takahashi and M. Nakamura (2008). "The expression of TNF-alpha converting enzyme at the site of ruptured plaques in patients with acute myocardial infarction." Eur J Clin Invest 38: 97-105. 272

Sauer, L., D. Gitenay, C. Vo and V. Baron (2010). "Mutant p53 initiates a feedback loop that involves Egr-1/EGF receptor/ERK in prostate cancer cells." Oncogene 29: 2628-2637. Schafer, B., A. Gschwind and A. Uldrich (2004). "Distinct ADAM metalloproteinases regulate G protein-coupled receptor-induced cell proliferation and survival." J Biolo Chem 279: 47929-47938. Schafer, B., A. Gschwind and A. Uldrich (2004). "Multiple G-protein-coupled receptor signals converge on the EGF receptor to promote migration and invasion." Oncogene(23): 991-999. Schneider, M. and E. Wolf (2008). "The epidermal growth factor receptor ligands at a glance." J Cell Physiol 218: 460-466. Schollman, C., R. Grugel, D. Tatje, J. Hoppe, D. Marme and H. Weich (1992). "Basic fibroblast growth factor modulates the nitogenic potency of the platelet-derived growth factor (PDGF) isoforms by specific up-regulation of the PDGF alpha receptor." J Biolo Chem 267: 1032-1039. Schwartz, S. (1997). "Smooth muscle cell migration in atherosclerosis and restenosis." J Clin Investigation 99: 2814-2817. Schwartzberg, P. (1998). "The many faces of Src: multiple functions of a prototypical tyrosine kinase." Oncogene 17(11Reviews): 1463-1468. Seifert, R., S. Coats, E. Raines and R. Ross (1994). "Platelet-derived growth factor receptor alpha subunit mutant and reconstituted cell lines demonstrate that transforming growth factor beta can be mitogenic through PDGF-A-chain-dependent and - independent pathways." J Biol Chem 269: 13951-13955. Shah, P. (2007). "Molecular mechanisms of plaque instability." Curr Opin in Lipidol 18: 492-499. Shamin, A., T. Pelzer, C. Grhe and L. Neyses (1999). "Induction of Egr-1 mRNA and protein by endothelian 1, angiotensin II and norepinephrine in neonatal cardiac myocytes." Molec and Cell Biochem 195: 11-17. Shanahan, C. and P. Weissberg (1999). "Smooth muscle cell phenotypes in atherosclerotic lesions." Curr Opin in Lipidol 10: 507-513. Shunichi, S. and K. Tsumiyama (2009). "Pathogenesis of reumathoid arthritis and c- Fos/AP-1." Cell Cycle 8(10): 1539-1543. Sibilia, M., R. Kroismaryr, B. Lichtenberger and A. Natarajan (2007). "The epidermal growth factor receptor: from development to tumorigenesis." Differentiation 75: 770-787. Silverman, E. and T. Collins (1999). "Pathways of Egr-1 mediated gene transcription in vascular biology." Am J of Pathol 154(3): 665-670. Skiles, J. and A. Gonnella (2001). "The design, structure and therapeutic application of matrix metalloproteinase inhibitors." Curr Med Chem 8: 425-474. Soond, S., B. Everson, D. Riches and G. Murphy (2005). "ERK-mediated phosphorylation of Thr735 in TNFalpha-converting enzyme and its potential role in TACE protein trafficking." J Cell Scien 118: 2371-2380. Soriano, P. (1994). "Abnormal kidney development and haematological disorders in PDGF receptor beta mutant mice." Genes Dev 8: 1888-1896. Sounni, N. and A. Noel (2005). "Membrane type-matrix metalloproteinases and tumor progression." Biochem J 87: 329-342. 273

Sternlicht, M. and Z. Werb (2001). "How metalloproteinases regulate cell behaviour?" Ann Rev Cell Dev Biol 17: 463-516. Stice, L., C. Vaziri and D. Faller (1999). "Regulation of platelet-derived growth factor signaling by activated p21-Ras." Frontiers in Bioscien 4: 72-86. Sundaresanan, M. (1995). "Requirement for generation of H2O2 for platelet-derived growth factor signal transduction." Scien 270(5234): 296-299. Suske, G. (1999). "The Sp family of transcription factors." Gene 238: 291-300. Suzuki, J., M. Iwai, H. Nakagami, L. Wu, R. Chen, T. Sugaya, M. Hamada, K. Hiwada and M. Horiouchi (2002). "Role of angiotensin II-regulated apoptosis through distinct AT1 and AT2 receptors in neointimal formation." Circ 106: 847-853. Swan, D. and W. McBride (1982). "Chromosomal mapping of sarcoma virus onc gene analogue in human cells." Proc Nat Acad Sci USA 79: 4691-4695. Tamura, R., J. Miyagawa and M. Nishida (2001). "Immunohistochemical localization of betacellulin, a member of epidermal growth factor family in atherosclerotic plaques from human aortae." Atheroscle 155: 413-423. Tamura, Y., F. Watanabe and T. Nakatani (1998). "Highly selective and orally active inhibitors of type IV collagenase (MMP-9 and MMP-2: N-sulfonylamino acid derivatives." J Med Chem 41(4): 640-649. Tan, N., V. Midgley, M. Kavurma, F. Santiago, X. Luo, R. Peden, R. Fahmy, M. Berndt, M. Molloy and L. Khachigian (2008). "Angiotensin II-inducible platelet-derived growth factor-D transcription requires specific Ser/Thr redidues in the second finger region of Sp1." Circ Res 102: e38-e51. Thiel, G., S. Mayer, I. Muleer, L. Sefano and Y. Roosler (2010). "Egr-1-A Ca2+-regulated transcription factor." Cell Calcium 47: 397-403. Thomas, W., Y. Brandenburger, D. Autelitano, T. Pham, H. Qian and R. Hannan (2002). "Adenoviral-directed expression of the type 1A Angiotensin receptor promotes cardiomyocyte hypertrophy via transctivation of the epidermal growth factor receptor." Circ Res 90: 135-142. Touyz, U., X. Wu, G. He, J. Park, X. Chen, J. Vacher, V. Rajapurohitam and E. Schiffrin (2001). "Role of c-Src in the regulation of vascular contraction and calcium signaling by angiotensin II in human vascular smooth muscle cells." J Hyperten 19(3): 441-449. Toyoda, H., T. Komurasaki, D. Uchida and S. Morimoto (1997). "Distribution of mRNA for human epriregulin, a differentially expressed member of the epidermal growth factor family." Biochem 326(1): 69-75. Tsai, J., L. Liu, J. Guan and W. Aird (2000). "The Egr-1 gene is induced by epidermal growth factor in ECV304 cells and primary endothelial cells." Am J Physiol Cell Physiol 279: C1414-1424. Ueda, M., A. Becker, N. Kasayuki, A. Kojima, Y. Morita and S. Tanaka (1996). "In situ detection of platelet-derived growth factor-A and factor -B chain mRNA in human coronary arteries after percutaneous transluminal coronary angioplasty." Am J Pathol 149(3): 831-843. Ustach, C., M. Taube, N. Hurst, K. Paavonen, I. Tajantie, M. Lohela, H. Wiig, K. Pajusola, U. Eriksson and K. Alitalo (2004). "PDGF-D induces macrophage recruitment, increased interstitial pressure and blood vessel maturation during angiogenesis." Blood 104: 3198-3204. 274

Utela, M., J. Lauren, E. Bergsten, X. Li, U. Eriksson and K. Alitalo (2001). "Chromosomal location, exon structure and vascular expression patterns of the human PDGF-C and PDGF-D genes." Circ Res 103: 2242-2247. Van der Gee, P. and T. Hunter (1994). "Receptor protein-tyrosine kinases and their signal transduction pathways." Ann Rev Cell Biol 10: 251-337. Vanhoutte, D. and S. Heymans (2010). "TIPMs and cardiac remodeling: "Embracing the MMP-independent-side of the family"." J Molec and Cell Cardiol 48: 445-453. Vassbotn, F., O. Havnen, C. Heldin and H. Holsmen (1994). "Negative feedback regulation of human platelets via autocrine activation of the platelet-derived growth factor alpha and beta." J Cell Biochem 269: 13874-13879. Vincenti, M. and C. Brinckerhoff (2007). "Signal transduction and cell-type specific regulation of matrix metalloproteinases gene expression: Can MMPs be good for you?" J Cell Physiol 213: 355-364. Virolle, T., A. Krones-Herzig and V. Baron (2003). "Egr-1 promotes growth and survival of prostate cancer cells." J Biolo Chem 278: 11802-11810. Wagsater, D., C. Zhu, H. Bjork and P. Ericksson (2009). "Effects of PDGF-C and PDGF-D on monocyte migration and MMP-2 and MMP-9." Atheroscl 202(2): 415-423. Wang, B., J. Chen, F. Santiago, M. Janes, M. Kavurma, B. Chong, J. Pimanda and L. Khachigian (2010). "Phosphorylation and acetylation of histone-3 and autoregulation by early growth response-1 mediates interleukin-1 beta induction of early growth response- transcription." Arterioscler Thromb Vasc Biol 30: 536-545. Wang, C., G. Sharma and B. Draznin (2006). "Early growth response gene-1 expression in vascular smooth muscle cells." Am J of Pathol 19: 366-372. Wang, X., T. Oka, F. Chow, S. Cooper, J. Odenbach, G. Lopaschuk, Z. Kassiri and C. Fernandez-Patron (2009). "Tumor necrosis factor-alpha-converting enzyme is a key regulator of agonist-induced cardiac hypertrophy and fibrosis." Hyperten 54: 575- 582. Webb, K., A. Henney and S. Anglin (1997). "Expression of matrix metalloproteinases and their inhibitor TIMP-1 in the rat carotid artery after balloon injury." Arterioscler Thromb Vasc Biol 17(9): 1837-1844. Westmarck, J., A. Seth and V. Kahari (1997). "Differential regulation of interstitial collagenase MMP-1 gene expression by ETS transcription factors." Oncogen 14(2651-60). White, S. and A. Newby (2002). "Gene therapy for all aspects of vein graft disease." J Card Surg 17(6): 549-555. Wilson, C., A. Ouellette, D. Satchel, T. Ayabe and Y. Lopez-Boado (1999). "Regulation of intestinal alpha-defensin activation by the metalloproteinase matrisylin in innate host defense." Scien 286: 113-117. Yamakana, Y., K. Hayashi and T. Komurasaki (2001). "EGF family ligand-dependent phenotypic modulation of smooth muscle cells through EGF receptor " Biochem Biophys Res Commun 281: 373-377. Yan, C. and D. Boyd (2007). "Regulation of matrix metalloproteinases gene expression." J Cell Physiol 211: 19-26. Yan, Y., N. Shrirakabe and Z. Werb (2002). "The metalloprotease Kuzbanian (ADAM10) mediates the transactivation of EGF receptor by G protein-coupled receptors." J Cell Biol(221-6). 275

Yang, X., M. Zhu, N. Sreejayan and J. en (2005). "Angiotensin II promotes smooth muscle cell proliferation and migration through release of heparin-binding epidermal growth factor and activation of EGF-receptor pathway." Mol Cells 20(2): 263-270. Yang, Y., S. Yuzaka and J. Schlessinger (2008). "Contacts between membrane proximal regions of the PDGF receptor ectoomain are required for receptor activation but not for receptor dimerization." PNAS 105(22): 7681-7686. Yoshida, T. and G. Owens (2005). "Molecular determinants of vascular smooth muscle cell diversity." Circ Res 96: 280-291. Yun, S., A. Dardik, M. Haga and A. Yamashita (2002). "Transcription factor Sp1 phosphorylation induced by shear stress inhibits membrane type 1-matrix metalloproteinase expression in endothelium." J Biolo Chem 277: 34808-34814. Zhang, Q., S. Shenoy, K. Khalili and D. Kaetzel (2005). "Receptor alpha activates PDGF-A gene transcription via interactions with a G-rich region, single stranded of the promoter." Gene 348: 25-32. Zhang, X. and Y. Liu (2003). "Suppression of HGF receptor gene expression by oxidative stress is mediated through the interplay between Sp1 and Egr-1." Int J Physiol Renal Physiol 284(6): F1216-1225. Zhang, Z., P. Oliver and J. Lancaster (2001). "ROS mediate tumor necrosis factor-alpha- converting enzyme-dependent ectodomain shedding induced by PMA." FASEB 15: 303-305. Zhou, W., B. Ibe and U. Raj (2007). "Platelet-activating factor induces ovine fetal pulmonary venous smooth muscle cell proliferation: role of EGRF transactivation." Am J Physiol Heart Circ Physiol 292: H2773-H2781. Zhuo, Y., M. Zhang and J. Lasky (2004). "Modulation of PDGF-C and PDGF-D expression during bleomycin-induced lung fibrosis." Am J Physiol Lung Cell Mol Physiol 286: 182-188. Zwerner, Y. and W. May (2001). "PDGF-C is an EWS/FLI induced transforming growth factor in Ewing family tumors." Oncogen 20: 626-633.